Sense and nonsense in psychology

a Pelican Original $1.25 1

Sense and Nonsense
in Psychology

H. J. Eysenck

PELICAN BOOKS

A3°5

SENSE AND NONSENSE IN PSYCHOLOGY

H.J. EYSENCK

Dr H J Eysenck obtained his Ph.D. degree in psychology at
London Univers.ty after school and university experiencT L
Germany, France, and England. Having worked as psychol"
gist at the war-tune Mill Hill Emergency Ho,™,,! T.
appointed Professor of Psychology a, Sft LondT
and Director of the Psychological Department at die InsSme
of Psychiatry (Maudsley and Bethlem Royal Hospital) He

timem *:umSrnyvCOUntriepS' ^ ^ ^ ^™
time in the U.S.A. as Vis.tmg Professor at the Universities of

Pennsylvania and California. Known mainly through hi

experunental researches in the field of personality he has

written some two hundred articles in technical journal a

well as several books including Dimensions of PersonXYhl

ScunUfic Study of Personality, The Structure of Human Prfonam

The Psychology of Politics, The Dynamics of Anxiety an^HystZ'

and I Perceptual Processes and Mental Illness. He is Editor-mSnef

I f.K IT"13' Bechamm Research and ^"Py- He advocates the
highest degree of scientific rigour in the design of psychological
expcnments and .very critical of much loofe thinkLg current
at present under the guise of 'psychology'. g

Digitized by the Internet Archive

in 2010 with funding from

Lyrasis Members and Sloan Foundation

http://www.archive.org/details/sensenonsenseinpe98seyse

SENSE AND NONSENSE
IN PSYCHOLOGY

By
H.J. EYSENCK

PENGUIN BOOKS

BALTIMORE • MARYLAND

Penguin Books Ltd, Harmondsworth, Middlesex, England

Penguin Books Inc., 3300 Clipper Mill Road, Baltimore 11, Md, U.S.A.

Penguin Books Pty Ltd, Ringwood, Victoria, Australia

First published 1957

Reprinted with revisions 1958

Reprinted i960, 1961, 1962, 1963, 1964

Copyright © H. J. Eysenck, 1957, 1958

Made and printed in Great Britain

by Richard Clay & Company, Ltd, Bungay, Suffolk

Set in Monotype Baskerville

This book is sold subject to the condition

that it shall not, by way of trade, be lent,

re-sold, hired out, or otherwise disposed

of without the publisher's consent,

in any form of binding or cover

other than that in which

it is published

CONTENTS

Editorial Foreword J

Introduction 1 1

part one: Borderlands of Knowledge

1. Hypnosis and Suggestibility 25

2. Lie Detectors and Truth Drugs 71

3. Telepathy and Clairvoyance 1 06

4. The Interpretation of Dreams 142

part two: Personality and Social Life

5. Can Personality be Measured? 175

6. Personality and Conditioning 231

7. Politics and Personality 265

8. The Psychology of Aesthetics 308
Recommended for Further Reading 340

Index 341

EDITORIAL FOREWORD

The building of a science is in many ways like the building of a
cathedral or the building of a city. It is rarely, if ever, a matter of
starting on a virgin site and of executing in detail a preconceived
plan. More often than not it is a matter of continuous alteration,
of repeated destruction and reconstruction. At any time the edifice
will display the styles of many ages.

This is certainly so with Psychology. The science so described is
an edifice which today has some of the features of the Parthenon,
and some of the features of the new Coventry Cathedral. If its
original architect could be named it would be Aristotle, but
the system incorporates elements deriving from the pre-Socratics
and elements from primitive thought; and like other ancient
foundations it embodies much picturesque but functionless
ornamentation.

It follows that a good psychologist must be a man of several
parts. He must be original and creative, but he must also be
critical and destructive. The needed qualities of mind are rarely
combined, and in general the progress of psychology depends upon
the kind of cooperation that consists in conflict. 'Bold creative
thinkers' (as we describe them when they have our approval) or
'armchair speculators' (as we describe them when they have not)
submit their creative thoughts to be manhandled and mauled by
carping destructive critics and by the manipulators of slide rules.
Never does a grand overall theory survive in the form in which it
was originally conceived. Rarely, if ever, is it utterly destroyed.
Something comes out of this conflict which neither the creative
thinker nor the destructive critic could have produced alone.

Sometimes the qualities of the destructive critic are combined
with those of the constructive thinker - as in the case of the author
of this book. Dr Eysenck can manipulate slide rules, but he can
also manipulate concepts. He has displayed these skills not only in
his many contributions to the research journals, but also in works
adapted to the needs and interests of the layman. Among the latter
is the earlier volume in this series on Uses and Abuses of Psychology.

The layman can be pardoned for his belief that there is an essen-
tial incompatibility between an interest in interesting things and
the capacity to discuss these things in a scientific way; and he can
be pardoned for his belief that there is an essential association of

8 Editorial Foreword

scientific ability with dry-as-dustiness. The author of this book is
an awkward negative case for those who would defend either
generalization. He certainly discusses interesting questions and he
discusses them in both a scientific and an interesting way. Of
course there are other scientists, equally expert and equally fair-
minded, who arrive at conclusions different from some of those
presented in this book. The author would certainly not be the last
to agree to this. But that is neither here nor there. What is im-
portant is that interesting and potentially explosive issues should
be matters of rational discussion. Sense and Nonsense in Psychology,
like The Uses and Abuses of Psychology, is a book which contributes
not only to entertainment and instruction but also to the diffusion
of a certain quality characteristic of the intellectual life of those
who live in civilized societies.

G. A. Mace
October 1956

Acknowledgements for permission to quote are
due to the authors and publishers of the follow-
ing two books:

Clark L. Hull. Hypnosis and Suggestibility.

New York: Appleton- Century-Crofts,

Inc., 1933.
F. E. Inbau & J. E. Reid. Lie Detection

and Criminal Interrogation. Third Edition.

Baltimore: The Williams & Wilkins Co.,

1953.

The author is indebted to Modern Experiments in
Telepathy by Soal and Bateman for many of the
criticisms of card-guessing on pages 1-228.

Science is nothing but trained and organized common-
sense, differing from the latter only as a veteran may
differ from a raw recruit: and its methods differ from
those of common-sense only as far as the guardsman's
cut and thrust differ from the manner in which
a savage wields his club.

The chess board is the world; the pieces are the pheno-
mena of the universe; the rides of the game are what we
call the laws of nature. The player on the other side is
hidden from us. We know his play is always fair, just,
and patient. But also we know, to our cost, that he
never overlooks a mistake or makes the smallest
allowance for ignorance.
*

T. H. HUXLEY

INTRODUCTION

In a sense, this book is a sequel to Uses and Abuses of Psycho-
logy, but it has been written to stand on its own feet. In the
first book I was concerned with problems of social usefulness
and the contribution which psychology can make to society,
and with the many unfounded claims which have been made
in its name and which have caused so many intelligent
critics to discount psychological contributions almost com-
pletely. In the present book I have taken a rather wider
field. There is no doubt that a great deal of nonsense is
talked, even by educated people, on a large number of sub-
jects, but it is unlikely that there is any subject where the ratio
of sense to nonsense is smaller than in matters psychological.
I have picked out a few fields in an attempt to sort out the
chaff from the wheat, and to put before the reader the facts
as well as the speculations, the sense as well as the nonsense,
which have grown up around these topics.

There is little obvious usefulness in most of the subjects
discussed, although it is not out of the question that, in due
course, hypnosis or dream interpretation, or even telepathy,
may reach a status where claims in that direction could be
substantiated. However, the discussion is certainly not
slanted in that direction; our interest has been purely in the
subject-matter as such; in the facts, and in the theories
woven around these facts in an attempt to explain them.

It is interesting to consider for a moment why there are so
many misconceptions about psychological topics. I vividly
remember an occasion, not so very many years ago, when
the Vice- Chancellor of one of our main universities, on see-
ing that on the occasion of a celebration dinner his wife had
been placed next to the Professor of Psychology at the table,
hurriedly changed the cards around, explaining that it
would not be quite proper for a lady to sit next to a psycho-
logist! Times may have changed a little since then, and a
few people may have learned that the aura of sex, licentious-
ness, and promiscuity which, in the popular mind, attaches

12 Introduction

to psychoanalysis cannot justifiably be transferred to the
innocent psychologist running his rats in the animal labora-
tory, or eliciting the laws of learning by conditioning univer-
sity students to give an involuntary eye-blink to the sound of
a buzzer. Yet, the great majority of people still have little
idea of what it is that the psychologist tries to do. An oft-
repeated reaction of the layman - and even more of the lay-
woman - on being introduced to a psychologist is the
stereotyped sentence, { Oh dear, I bet you'll know all that's
in my mind!' While the psychologist as a private person
would occasionally like to know what is in the other person's
mind, he would certainly not regard this as his professional
job. What he is trying to do is simply to discover the laws
according to which human and animal behaviour takes
place; the laws of learning, of memory, of emotion; the rules
governing the development of motor and perceptual skills ;
the ways in which intellectual abilities develop and grow;
and the laws of social and interpersonal relationships. This
is quite a full programme, and by and large surely a sensible
one ; there is no black magic in all this, but simply the appli-
cation of the usual scientific methods to a relatively complex
and difficult subject.

In practice, most people show a rather odd and interest-
ing ambivalence to a programme such as this. On the one
hand, there is considerable verbal acclaim and approval.
Most people realize that our knowledge of the physical
world has vastly outstripped our knowledge of mankind,
and that unless we can find some way of correcting this
balance there may be very little left of mankind to study.
This view, indeed, has become so much a commonplace that
one feels almost ashamed of putting it down in black and
white again. Yet, curiously enough, this almost universal
verbal approval is not in any way reflected in action. The
total amount of money spent on all the social sciences in this
country is less than one thousandth of what is spent on the
physical sciences; indeed, when we take into account the
money spent by industry, the ratio becomes much more un-
favourable still. In the whole country there are, at most, one

Introduction 13

dozen small, under-staffed, and under-developed depart-
ments of psychology lacking mostly even the elementary
laboratory facilities required for teaching. On the practical
side, then, it is reasonably safe to say that the lofty aspira-
tions voiced by our so-called leaders of thought remain pre-
cisely that - aspirations - and that their translation into
reality is put off until the Greek calends.

It often appears as if people, while paying lip-service to
the desirability of basing conclusions on facts, prefer the
meretricious sesquipedalianism of the kind offered by the
late C.E.M.Joad and his many successors. Psychological
topics of which there is some factual knowledge in existence
are often dealt with by the BBC, but they are nearly always
dealt with by philosophers, zoologists, mathematicians,
journalists, theologians, or unspecified dons of ancient vin-
tage, who clearly do not know even that there is any factual
knowledge available, and who speculate and pontificate to
their hearts' content in a manner they would never dare to
assume in their own specialty.

This is only one example of what appears to be an almost
universal belief to the effect that anyone is competent to dis-
cuss psychological problems, whether he has taken the
trouble to study the subject or not, and that while every-
body's opinion is of equal value, that of the professional
psychologist must be excluded at all costs because he might
spoil the fun by producing some facts which would com-
pletely upset the speculation and the wonderful dream-
castles so laboriously constructed by the layman.

This tendency is most noticeable, of course, when the facts
provided by the psychologist have a bearing on political
matters. It is then that we find the finest flower of nonsense.
Care is, naturally, always taken that the discussion should be
pursued by people not contaminated in any way by factual
knowledge, and motivated almost exclusively by a desire to
score party points. One example may illustrate the sort of
thing that is happening almost every day. In Uses and Abuses
of Psychology, I described in some detail the construction of
intelligence tests, the methods of validating them, and their

14 Introduction

usefulness in selection. In recent years there has been a con-
siderable degree of criticism of selection, particularly at the
1 1 -plus level, and it is worth-while to have a look at the
most common of these objections. It is argued, for instance,
that many children are so anxious on the occasion of the test
that they do not do themselves justice. It is argued again
that coaching produces a considerable effect on test scores,
and that therefore the results are worthless. It is argued that
the age at which selection takes place is too early in the life
of a child, and that either no selection should take place at
all, or that some form of 'streaming' should be introduced,
with the constant possibility of changing the child from one
stream to another. Whatever the objection, the argument
usually ends by saying that intelligence tests are either use-
less or actively malfeasant in their influence.

Much of this argument arises from a very praiseworthy
desire to give all children an equal chance, and to act on the
principle that all men are created equal. Unfortunately,
the facts make it quite certain that all men are not created
equal, and that heredity clearly discriminates between the
bright and the dull. In fact, the more we equalize the oppor-
tunities of education, the greater will be the influence of
heredity in determining the final intellectual status of each
child. This point is often argued by Communists, whose
strange dogma seems to favour the conception of human be-
ings as emerging with perfect uniformity from some con-
ceptual conveyor belt, but no such view can be maintained
for one moment in the face of the evidence.

Given innate differences in ability, we must next stress a
fact which again has ample experimental foundation,
namely that teaching is much more effective when the mem-
bers of the class are of relatively uniform ability. It has often
been shown that to teach the same material to a group con-
taining bright, average, and dull members takes consider-
ably longer and is considerably less effective than teaching
the same material to a group consisting only of bright, or
average, or dull members. It might, at first sight, seem sur-
prising that it should be easier to teach a dull class than one

Introduction 1 5

of average ability, but the answer, of course, is that with a
dull class the teacher can adapt himself and his methods to
the level of his students, while with a class of average ability,
containing both bright and dull students, the same method
of teaching simply will not be applicable to all members of
the class, so that time-wasting duplication is necessary. Un-
less, therefore, we want to have highly inefficient teaching,
we must segregate - quite apart from the fact that certain
subjects at a reasonably advanced level are quite unsuitable
for children with low I.Q.s.

Given that for these, as well as for many other reasons,
some form of segregation is desirable, is eleven too early an
age at which to estimate children's abilities ? Follow-up studies
have usually shown that there is very little overlap between
the abilities of children in grammar and in modern schools,
even when they reach the age of fifteen or sixteen, so that
prediction in the vast majority of cases appears to have been
remarkably accurate. What, then, about the criticisms of
coaching and examination nerves? Both are justified, but
they are criticisms, not of psychological tests and psychologi-
cal theory, but of political and social pressures which pre-
vent the best use being made of available knowledge. There
is ample evidence that coaching produces a rise in I.Q,. of
about ten points or so, which is quite a considerable increase.
It has also been shown, however, that simply giving chil-
dren a few hours practice in doing intelligence tests has
much the same effect as coaching, and brings them to a
point beyond which no coaching and no practice is found to
increase their scores. The answer to the coaching problem,
therefore, is quite a simple one. All children should be given
five hours of practice on tests of intelligence before the
crucial test is taken. This would counteract the effects of any
coaching they would be given additionally, and give them
all the same start. At the same time, this would reduce ner-
vousness due to having to face an entirely new experience.

It would be preferable in many ways to have these five
hours of testing spaced throughout the child's school period
from six to eleven so that some knowledge could be gained

1 6 Introduction

at the same time of his intellectual growth pattern, because
this might be of considerable value in making any predic-
tions. Also it would reduce the importance of any particular
single test, and would thus again have some effect in reduc-
ing examination nervousness. Children who did not come
up to their expected level could be re-tested individually to
determine whether this failure was due to nervousness or
other extraneous causes. There are other ways in which the
experienced psychologist could help education committees
to overcome the difficulties involved in the administration of
testing programmes and obviate the criticisms mentioned.
Why, then, is nothing of this kind done?

The reason is a very simple one. If it costs gaf per child to
give one test of intelligence and score it, then it would cost
almost 4s to give five tests. The decision that a child's future
happiness is worth only gd rather than 3 s gd is not made by
the psychologist, but by the general public through their
elected members of local government bodies. All one can
say is that for gd the public is getting quite incredibly good
value for its money. To have a fairly superficial examination
of a car carried out costs several pounds. A reasonably com-
plete health examination costs at least as much. To get a
child's intelligence investigated for gd can hardly be con-
sidered an extravagant expenditure. The degree of accuracy
of this measurement, of course, is directly dependent on the
amount of money spent on it. The higher the degree of ac-
curacy required, the more money will have to be spent on
the examination. During my recent visit to California I was
shown a large series of laboratories constructed at a cost of
several million dollars for the sole purpose of getting a few
thousandths of one degree nearer to absolute zero tempera-
ture ! Our society is willing to pay large sums of money like
this for a slight increase in the accuracy of physical measure-
ment, but it is content with an expenditure of gd a child in
the measurement of a psychological variable of great im-
portance to both the child and society.

It is not my purpose in making this comparison to say
whether this social decision is right or wrong. It is my pur-

Introduction 1 7

pose rather to point out some of the facts of the situation,
and to show that if the accuracy of the measurement per-
formed of a child's ability at the age of 1 i-plus is not all it
might be, and is subject to various types of criticism, then it
is not the psychologist who should be blamed, but society it-
self. By refusing to spend money on research, or refusing to
implement the findings of private research because of a
slight increase in cost, society has made a decision which
may or may not be a correct one, yet discussion in the Press
heaps all the resulting blame on the head of the psychologist
for not performing miracles at the price of gd, and fails to
put before the public the facts of the situation so that it may
have a chance to reconsider its verdict.

One of the reasons why popular discussions on psycho-
logical matters are so uninformed is perhaps the fact that
psychology, being a science, uses methods and discussions of
a highly technical nature. In particular, it uses statistical
methods, and most people have developed a strong avoid-
ance reaction to statistics. This reaction is both interesting
and odd because it seems to be rather one-sided. A discus-
sion of odds in horse-racing, or permutations and combina-
tions in football pools, is no less statistical for being popular
and widespread, yet when the psychologist attempts to em-
ploy statistical methods in relation to serious subjects he en-
counters the usual obscurantist notion that you can prove
anything with statistics. In a sense, this is true. You can
'prove' anything with statistics, but only to a person who is
completely ignorant of elementary statistical notions. In
fact, faulty arguments, which the statistician would recog-
nize as such immediately, often pass muster in the popular
Press because they are disguised in such a way that their
statistical nature is not recognized.

Let me give two examples of this : It is well known that
about half the hospital beds in Britain are occupied by
mental patients. On quite a number of occasions I have seen
journalists argue from this as if half the medical patients in
the country were suffering from mental disorders. Super-
ficially, this may not seem a statistical point at all, yet the

1 8 Introduction

fallacy quite definitely is a statistical one. The equation of
'half the beds' with 'half the patients' would be correct only
if we make the assumption that the length of time a patient
stays in his bed is the same for mental and physical dis-
orders. It is quite obvious that if, on the average, the patient
suffering from a physical disorder stays in bed for two days
and is then discharged, whereas the mental patient on the
average stays for twenty years, then the number of patients
suffering from physical disorders is 3,650 times as great as
the number of patients suffering from mental disorders.
Now, it is a fact that a large number of mental patients stay
in hospital for very long periods, whereas patients suffering
from physical disorders tend, in the great majority of cases,
to stay only a few days. Unless the exact proportions are
known it is impossible to argue from the proportion of beds
to the proportion of patients.

Another example may be taken from the controversy on
the causes of recidivism, i.e. the question of why certain
offenders return to prison again and again. I was visiting
Alcatraz, the famous American prison located on an island
in San Francisco Bay, where the most serious offenders in
the United States are punished. This prison is one of the
harshest and best guarded in the world, and it is claimed
that no one has ever succeeded in escaping from it. On this
visit I happened to be accompanied by a few penologists and
sociologists whose strong anti-statistical bias emerged during
the boat trip across the Bay. We were shown round the
prison by one of the deputy governors who had heard part
of this argument regarding the use of statistics. He winked at
me when the discussion turned to recidivism, and told the
group that Alcatraz had practically stamped out recidivism
because, of all the prisons in the country, it was right at the
bottom of the list with respect to the number of persons re-
turning to crime. This very much impressed his listeners,
who kept discussing the possible reasons for this pheno-
menon for quite a long time. It did not occur to them that
as most of the prisoners in Alcatraz are there for life, or for
a very long period, they would have very little chance of

Introduction 19

ever committing a crime again, simply because very few had
ever been released! This may seem an elementary point,
hardly worthy of being called statistical, but it is precisely
the kind of point so often missed by the unsophisticated
journalist or reader who attempts to dismiss the modern
logic of probability with a few damning phrases.

Sometimes society's reaction to psychological research
appears to be not only unenthusiastic, but actively male-
volent. As an example, let me take the recent inquiry
into the effects of television. There was considerable
social interest in this matter, and hundreds of people voiced
their opinions and attitudes in the complete absence of any
known facts. Some deplored the effects of television on chil-
dren, saying that it kept them indoors, kept them from read-
ing, made them incapable of initiative and enterprise, and
had a great variety of other disastrous consequences. Others
countered this by saying that it broadened their horizon,
made them more knowledgeable, and kept them from doing
other things which would be even less acceptable from the
point of view of society. Now here, clearly, is a point where
factual information is absolutely essential, and, indeed, the
papers clamoured for such information to be supplied.

Finally, one of our best-known social psychologists was
appointed by a Research Foundation to carry out and direct
a research into this problem. Numerous questionnaires,
diaries, and other instruments of research were circulated to
thousands of children forming a well-selected sample of the
school population. Out of this very large number, one or
two, in talking about the questionnaires to their parents,
giving an extremely garbled version of some of the ques-
tions, caused their parents to complain, and these com-
plaints were prominently featured in the national Press.
Certain points about the treatment by the Press of this mat-
ter are of considerable interest. The investigators had espe-
cially asked the newspapers not to print any of the actual
questions used because doing that would invalidate the
further use of the questionnaires containing these questions.
Few of the papers complied. Not only that, but what most

20 Introduction

of them printed were not the actual questions but the
garbled version of the questions obtained at third or fourth
hand from the parents of children who had answered the
original questionnaires. These questions, inaccurately
quoted and taken out of their context, bore very little rela-
tion to those which formed part of the actual inquiry. When
a Press conference was held and the correct questions were
given to the papers, most persisted in printing the garbled
version and did not print the corrections, nor did they
apologize.

This is the bare outline of only some of the reactions en-
countered in the course of this work, and the obvious hos-
tility and factual carelessness characterizing the Press treat-
ment of this issue is only too typical of what is happening
every day in journalistic discussions of psychological issues.
It is hardly to be wondered at that the amount of nonsense
talked about psychological matters is as great as it is.

It should not be thought, however, that the purveyors of
nonsense in this field are entirely to be found in the ranks of
non-psychologists. Alas, many members of the psychological
fraternity - and particularly the psychological sorority! -
have been equally guilty in this respect. It is hardly neces-
sary to document this assertion in any detail, as few people
are in any doubt about the facts of the matter. Let me illus-
trate my contention with just one example, namely, that of
the bringing up of children. The psychological literature on
this subject has always been one of the most amusing and
unedifying examples of faddism in the history of civilization.
During one decade all the books say the children should be
fed only at stated intervals, so that it is a crime for the
mother to feed her baby at five to one, while she has all the
blessings of science when she feeds him at one o'clock. Soon
the pendulum swings and everyone is instructed to feed her
children on demand, regardless of the time of day or night.
Parents recalcitrant to obey the currently fashionable de-
mands are threatened that their children will develop all sorts
of psychological disorders of a neurotic or psychotic kind if
these alleged psychological laws are not punctiliously obeyed.

Introductioji 2 1

An interesting point about this controversy, as well as
about so many other rules laid down on the bringing-up of
children, is that at no time is any factual evidence of an ex-
perimental kind brought forward to support the views
favoured by the text-book writers. Occasionally anecdotal
examples are given of terrible things which happen to cer-
tain children when brought up according to the non-
favoured rules and dictates, but the possibility that these
children might have suffered quite the same fate, regardless
of their upbringing, is never even canvassed.

Similarly absurd are the general notions current about the
later upbringing of children. Advocates of laisser /aire and
excessive leniency are on the left ; those of ' spare the rod,
spoil the child' on the right. Both schools have no hesitation
in telling the parent what to do, while neither appears to be
conscious of the fact that their precepts are based on un-
verified theories and idle speculation and that, in the ab-
sence of any factual knowledge regarding the consequences
of certain types of upbringing, the giving of advice is not far
removed from sheer impudence.

Quite generally, the notion that any set of precepts can be
useful as a general guide to human behaviour is one of the
finest exhibits in our museum of nonsense. The type of up-
bringing suitable for a child which forms conditioned re-
flexes easily might not be at all suited to a child who forms
conditioned reflexes only with difficulty; we shall discuss
this point at some length in a later chapter. Rules which
apply to the bright child do not necessarily apply to the dull
child. Upbringing suitable for the emotionally unstable
child may be quite unsuitable for the emotionally stable
child, and vice versa. Much painstaking research will be re-
quired before we can even begin hesitantly to give advice in
this field, which is anything but the projection of our own
prejudices and emotions. We may regret that times have not
advanced far enough to enable us to answer the many ques-
tions and problems in this field, but a recognition of the facts
of our ignorance is a more likely precursor to knowledge
than the pretence of omniscience.

22 Introduction

What is true of child upbringing is, a fortiori, true of such
complex social problems as war and peace, industrial un-
rest, and delinquency. Many people still have a pathetic
faith that if only everyone could be psychoanalysed, wars,
strikes, and crime could be wiped out as if by some magic
wand. Alack, the most likely effect, as far as the evidence in-
dicates, would be to leave matters precisely where they are
now, neither better nor worse. Science does not peddle vade-
mecums of this kind, and no credence should be given to
those who pretend to be able to solve all our problems by
some simple form of magic.

Why, it may be asked, is society so ambivalent in its atti-
tude towards the scientific factual and experimental study
of psychological phenomena, and why does it so readily
accept the unfounded claims for all sorts of psychological
panaceas? It is difficult to know for certain, but one likely
possibility is perhaps this: politicians and journalists are
used to what Humpty Dumpty called a knock-down argu-
ment, i.e. a purely verbal discussion, depending entirely on
the manipulation of symbols mostly quite unrelated to real-
ity. Having once tasted the exhilaration of being accepted
as * expert' on social and psychological matters because of
their ability to manipulate linguistic symbols, such people
are naturally not too eager to see factual research expose the
half-truths and vague generalizations wThich are their stock-
in-trade. This would account for the fact that while verbally
praising the efforts of the experimentalist, they yet put every
obstacle in the way of his successfully completing his investi-
gations, and pay no attention to his conclusions once the
experiment is finished. They are much more at home with
the non-factual, semantic arguments of the psychological
crusaders who have a system to sell and who therefore meet
them on their own ground. No need to be bothered here by
hard facts; acceptance and rejection can proceed in terms of
emotion and prejudice rather than in terms of proof and
disproof.

So much, then, for a brief introduction to the general
topic of this volume. The arrangement of the different sub-

Introduction 23

jects and their classification have caused me a good deal of
thought, particularly the attempt to find general terms to
denote the two main parts of the book. My difficulties here
are very similar to those faced by the Royal Commission
investigating prostitution, one of whose first tasks was to find
a collective noun for its objects of study. The first suggestion
made, {A jam of tarts', was rejected because of the objec-
tions of one of the Bishops, who thought it a trifle vulgar.
The second suggestion, 'A novel of Trollop eV, was received
with more favour, but was considered a trifle on the literary
side. The third suggestion, 'An anthology of pros', was
finally rejected for the same reason, and agreement finally
reached on the fourth and last term proposed, namely, 'A
fanfare of strumpets'. I was unable to match this inventive-
ness in relation to my own problem, but hope that the two
descriptive phrases finally chosen - Borderlands of Knowledge
and Personality and Social Life - will give some impression to
the reader of the subjects dealt with in the two parts of this
book.

PART ONE

BORDERLANDS OF KNOWLEDGE

1

HYPNOSIS AND SUGGESTIBILITY

The theory and practice of hypnosis is certainly a topic
which must have a place of honour in any book dealing
with sense and nonsense in psychology. Very few topics in
the whole history of mankind can have given rise to so many
absurdities, misunderstandings, and misconceptions. From
the very beginning the study of hypnosis has been tied up
with fantastic conceptions like animal magnetism, the in-
fluence of the stars, and similar tarradiddle. Even nowadays
popular conceptions of hypnosis are extremely confused,
and journalistic reports in newspapers have done but little
to clarify the issues involved.

Most of the experimental work done in this field in recent
years has been concerned with clearing up the errors com-
mitted by the early hypnotists, and consequently it is neces-
sary to provide the reader with a brief historical circum-
spect. This inevitably begins with the rather mysterious
personality of Franz Anton Mesmer, born in 1 733 in the little
Austrian village of Isnang, near Lake Constance. Mesmer
was intended for the priesthood and went to a monk's school
until the age of fifteen. He decided against the Church, be-
came a law student, but finally changed over to medicine. He
was awarded his degree at the age of thirty- two for a paper
1 concerning the influence of the planets on the human body \
This dissertation contained the first mention of his views on
animal magnetism, a theory which he was to amplify later in
a famous set of twenty-seven propositions. In essence, these
contained the main points of his teaching. According to them:

25

26 Part One • Borderlands of Knowledge

A responsive influence exists between the heavenly bodies, the
earth, and animated bodies; a fluid, universally diffused, is the
means of this influence. Experiments show that there is a diffusion
of matter subtle enough to penetrate all bodies without any con-
siderable loss of energy; its action takes place at a remote distance,
without the aid of any intermediary substance. It is, like light,
increased and reflected by mirrors. Properties are displayed,
analogous to those of the magnet, particularly in the human body.
This magnetic virtue may be accumulated, concentrated, and
transported. These facts show, in accordance with the practical
rules I am about to establish, that this principle will cure nervous
diseases directly, and other diseases indirectly. By its aid, the
physician is enlightened as to the use of medicine, and may render
its action more perfect and direct salutary crises so as to com-
pletely control them.

Thus, animal magnetism was held to be a kind of im-
palpable gas or fluid and its distribution and action were
supposed to be under the control of the human will. Not
only could this strange fluid be reflected by mirrors, but it
was also supposed to be capable of being seen. Trained
somnambulists were supposed to behold it streaming forth
from the eyes and hands of the magnetizer, though they ap-
pear to have disagreed as to whether its colour was white,
red, yellow, or blue! To the modern mind all this, of course,
seems nothing but a rigmarole of nonsense, but one posi-
tive assertion will be noted, namely Mesmer's allegation
cthat these principles will cure nervous diseases'. There ap-
pears to be little doubt that Mesmer did, in fact, produce
what were considered to be miraculous cures. Here, for in-
stance, is a publicly sworn statement by Charles du Hussey,
a Major of Infantry and Knight of the Royal Military Order
of St Louis.

After four years of useless treatment by other doctors, I con-
sulted Mesmer. My head was constantly shaking and my neck was
bent forward. My eyes protruded and were considerably inflamed.
My back was almost completely paralysed and I could only speak
with difficulty. I laughed involuntarily and for no obvious reasons.
My breathing was difficult and I suffered from severe pain between
the shoulders and from constant tremors. I staggered when I walked.

Hypnosis and Suggestibility 27

Du Hussey was then treated by Mesmer and went through
a series of strong emotional crises, 'ice coming from my
limbs, followed by great heat and foetid perspiration'. He
concludes his statement, 'Now, after four months, I am
completely cured'.

Mesmer's methods were certainly unorthodox, and his in-
terest in occult matters generally did not endear him to the
medical profession in Vienna. He took the precaution of
marrying a rich widow twelve years older than himself, thus
ensuring that his experiments would not be disrupted by
lack of money. However, his undoubted successes in curing
patients whom medical opinion had considered incurable
increased the detestation in which he was held by orthodox
physicians, and finally, when he was on the point of
curing the blindness of a very highly placed young girl who
had been given up by the great specialists of the time, ortho-
doxy succeeded in having him excommunicated from
Vienna and he went to Paris.

He was now forty-five years of age, and almost from the
beginning of his residence achieved a remarkable triumph. It
became fashionable to have ' nervous complaints ' and to have
these treated by Mesmer, very much in the same way that
it has become fashionable in the United States nowadays
to have some form of neurosis and be psychoanalysed. (The
average upper-middle-class inhabitant of New York, Boston,
Los Angeles, or Kansas City would be considered as much
out of things if he were unable to talk about his 'psycho-
analyst' as would his Parisian counterpart who could not
boast a mistress. The American habit is probably no less
expensive, certainly more virtuous, but rather less pleasur-
able, than the French. It probably does little harm, on the
whole, but equally it probably does little good.)

Indeed, pressure of work became such that Mesmer was
forced to introduce group therapy, very much in the way
that modern psychiatry has been forced in a similar direc-
tion. His clinic appears to have been a remarkable place in-
deed. Treatment was given in a large hall which was dark-
ened by covering up the windows. In the centre of this room

28 Part One • Borderlands of Knowledge

stood the famous baque, a large open tub about a foot high
and large enough to permit thirty patients to stand around
it. It was filled with water, in which had been placed iron
filings, ground glass, and a variety of bottles arranged in a
symmetrical pattern. The tub was covered with wooden
slats provided with openings through which jointed iron
rods projected. The patients applied these rods to their
various ailing parts, thus allowing the healing forces of ani-
mal magnetism to act. Patients were enjoined to maintain
absolute silence, and throughout the session plaintive music
was played by an orchestra hidden from view. At the psycho-
logical moment, Mesmer himself would appear in a brilliant
silk robe. He would move among the patients, fixing them
with his eyes, passing his hands over their bodies, and touch-
ing them with a long iron wand.

There can be no doubt that very many patients con-
sidered this treatment as beneficial, and claimed to have
been cured by Mesmer, again very much as modern
patients treated by psychoanalysis sometimes claim to have
been benefited by that technique. It is difficult nowadays to
tell whether Mesmer was a conscious charlatan who ex-
ploited the suggestibility of his patients, or whether he
seriously believed in the scientific truth of his hypothesis.
Bernheim speaks about the discredit which the interested
charlatanism of Mesmer threw upon his practices, but Moll,
who is perhaps the best-known and best-informed writer of
the history of hypnosis, is less severe. This is what he has to
say:

I do not wish to join the contemptible group of Mesmer's pro-
fessional slanderers. He is dead, and can no longer defend himself
from those who disparage him without taking into consideration
the circumstances or the time in which he lived. Against the uni-
versal opinion that he was avaricious, I remark that in Vienna, as
well as later in Morsburg and Paris, he always helped the poor
without reward. I believe that he erred in his teaching, but think it
is just to attack this only, and not his personal character. Let us
consider, however - for I deem it right to uphold the honour of one
who is dead - more closely in what his alleged great crime con-

Hypnosis and Suggestibility 29

sisted. He believed in the beginning that he could heal by means of
a magnet, and later that he could do so by a personal indwelling
force that he could transfer to the baque. This was evidently his
firm conviction and he never made a secret of it. Others believed
that a patient's mere imagination played a part, or that Mesmer
produced his effects by some concealed means. Then, by degrees,
arose the legend that Mesmer possessed some secret by means of
which he was able to produce effects on people, but that he would
not reveal it. In reality the question was not at all a secret pur-
posely kept back by him, since he imagined that he exercised some
individual force. Finally, if he used this supposititious force for the
purpose of earning money, he did nothing worse than do modern
physicians and proprietors of institutions who likewise do not fol-
low their calling from pure love of their neighbour, but seek to
earn their own living, as they are quite justified in doing. Mesmer
did not behave worse than those who nowadays discover a new
drug, and regard the manufacture of it as a means of enriching
themselves. Let us be just and cease to slander Mesmer, who did
only what is done by the people just mentioned, against whose pro-
cedure no one raises a word of protest, even when the drugs they
extol possess no therapeutic properties whatever.

Six years after his arrival in Paris, a dispute arose between
Mesmer and some of his disciples about the right to give
public lectures, revealing his supposed secret, and the
French Government intervened by appointing a Commis-
sion to investigate the truth of his claims. They were par-
ticularly interested in the so-called 'Mesmeric crises', an
example of which we have already encountered in con-
nexion with Major du Hussey. They ran a series of con-
trolled experiments of the kind which Mesmer himself
should have carried out before making any claims. This is
part of their report:

The commissioners were particularly struck by the fact that the
crises did not occur unless the subjects were aware that they were
being magnetized. For instance, in the experiments performed by
Jumelin, they observed the following fact. A woman, who appeared
to be a very sensitive subject, was sensible of heat as soon as
Jumelin's hand approached her body. Her eyes were bandaged,
she was informed that she was being magnetized, and she

30 Part One • Borderlands of Knowledge

experienced the same sensation, but when she was magnetized with-
out being informed of it, she experienced nothing. Several other
patients were likewise strongly affected when no operation was
taking place, and experienced nothing when the operation was
going on.

No wonder the commissioners came to the conclusion that
they had ' demonstrated by decisive experiments that imag-
ination apart from magnetism produces convulsions, and
that magnetism without imagination produces nothing.
They have come to the unanimous conclusion with respect
to the existence and utility of magnetism, that there is noth-
ing to prove the existence of the animal magnetic fluid; that
this fluid, since it is non-existent, has no beneficial effect;
that the violent effects observed in patients under public
treatment are due to contact, to the excitement of the imag-
ination, and to the mechanical imitation which involun-
tarily impels us to repeat that which strikes our senses. '

At about the same time the Royal Society of Medicine
made a very similar report to the effect that 'from a cura-
tive point of view animal magnetism is nothing but the art
of making sensitive people fall into convulsions'. These re-
ports finished Mesmer's career as a mental healer and he
left France shortly afterwards.

There is much in Mesmer's story which has interesting
parallels to the present time. I shall not take these up, how-
ever, until we have learned a little more about the nature of
hypnosis, and will instead go on to point out one fact which
has not even been commented on, although it is rather
curious. The terms 'mesmerize' and 'hypnotize' have be-
come quite synonymous, and most people think of Mesmer
as the father of hypnosis, or at least as its discoverer and first
conscious exponent. Oddly enough, the truth appears to be
that while hypnotic phenomena had been known for many
thousands of years, Mesmer did not, in fact, hypnotize his
subjects at all. Some of his patients appeared to have had
spontaneous hysterical convulsions and similar emotional
upheavals and abreactions, but there is no account in his
work or that of his followers of genuine hypnotic phenomena.

Hypnosis and Suggestibility 3 1

We thus start with the somewhat unusual finding that the
father of hypnosis never, in fact, hypnotized anybody, was
not familiar with the phenomena of hypnosis, and would
have had no place for them in his theoretical system. It is
something of a mystery why popular belief should have
firmly credited him with a discovery which in fact was made
by others.

The first to have induced the sleeping trance, which con-
stitutes the essential part of hypnosis, appears to have been a
pupil of Mesmer, the Marquis de Puysegur. In trying to in-
duce Mesmer's usual hysterical convulsion in a young shep-
herd, Victor, by the use of the magnetizing method, the
Marquis found that Victor had fallen into a quiet sleeping
trance from which he did not awake for quite a long time,
and which he was unable to recall after he had woken up.
This sleeping or trance condition, with the amnesia or for-
getting of all that occurred during the trance state following
it, attracted a good deal of attention, and very soon others
reported a variety of hypnotic phenomena, such as positive
hallucinations, i.e. seeing things not in fact there; negative
hallucinations, i.e. being blind to things actually present;
and anaesthesia, or failure to feel a touch on the skin ; anal-
gesia, i.e. an insensitivity to pain; and post-hypnotic sugges-
tion, or the tendency to carry out suggestions made under
hypnosis even after the hypnosis was terminated. We shall
discuss these phenomena in some detail a little later. First of
all, let us consider the method of producing the hypnotic
trance.

There are many methods of producing hypnosis ; indeed,
almost every experienced hypnotist employs variations
differing slightly from those of others. Perhaps the most
common method is something along these lines. The hypno-
tist tries to obtain his subject's co-operation by pointing out
to him the advantages to be secured by the hypnosis, such
as, for instance, the help in curing a nervous illness to be
derived from the patient's remembering in the trance cer-
tain events which otherwise are inaccessible to his memory.
The patient is reassured about any possible dangers he

32 Part One • Borderlands of Knowledge

might suspect to be present in hypnosis, and he may also be
told (quite truthfully) that it is not a sign of instability or
weakness to be capable of being put in a hypnotic trance,
but that, quite on the contrary, a certain amount of intelli-
gence and concentration on the part of the subject is abso-
lutely essential.

Next, the subject is asked to lie down on a couch, or sit in
an easy-chair. External stimulation is reduced to a mini-
mum by drawing the curtains and excluding, as far as pos-
sible, all disruptive noises. It is sometimes helpful to
concentrate the subject's attention on some small bright
object dangled just above eye-level, thus forcing him to look
slightly upwards. This leads quickly to a fatigue of the eye-
muscles, and thus facilitates his acceptance of the suggestion
that he is feeling tired and that his eyes are closing. The
hypnotist now begins to talk to the subject in a soft tone of
voice, repeating endlessly suggestions to the effect that the
subject is feeling drowsy, getting tired, that his eyes are clos-
ing, that he is falling into a deep sleep, that he cannot hear
anything except the hypnotist's voice, and so on and so
forth. In a susceptible subject, a light trance is thus induced
after a few minutes, and the hypnotist now begins to deepen
this trance and to test the reactions of the subject by giving
suggestions which are more and more difficult of execution.
Thus, he will ask the subject to clasp his hands together, and
tell him that it is impossible for him to separate his hands
again. The subject, try as he may, finds, to his astonishment,
that he cannot in actual fact pull his hands apart. Successful
suggestions of this kind are instrumental in deepening the
hypnotic trance until, finally, in particularly good subjects,
all the phenomena which will be discussed presently can be
elicited.

This, very briefly, is the routine method of establishing a
trance. It is very difficult to know just which of the elements
mentioned are really important. I have established a deep
hypnotic trance in a subject by quietly and soothingly say-
ing to him, * Monday, Tuesday, Wednesday, Thursday,
Friday, Saturday' in endless repetition. During the war, a

Hypnosis and Suggestibility 33

friend of mine had to hypnotize a French soldier who was
suffering from shell-shock. As the soldier spoke no English,
it was necessary to convey the suggestions in French. Un-
fortunately, the hypnotist was no great linguist and, to my
horror, he kept telling the subject, not that his eyes were
closing, but that his nostrils were! However, this made no
difference to the subject, who was soon in a deep trance, re-
gardless of what must, to him, have been a very odd type of
suggestion indeed. It does not seem, therefore, that the con-
tent of the suggestion is as important as one might have
thought at first glance.

On the other hand, the content of the suggestion does ap-
pear to be of some relevance. Here is an example quoted by
one of the best-known American hypnotists, who found
great difficulty in producing a deep trance in one of his sub-
jects. After several hours of intensive effort on the part of the
hypnotist, the subject timidly inquired if she might be
allowed to advise on technique, even though she had no
previous experience with a hypnotist. This is the advice she
gave: 'You are talking too fast on that point; you should say
that very slowly and emphatically and keep repeating it.
Say that very rapidly and wait a while and then repeat it
slowly, and please pause now and then to let me rest, and
please don't split your infinitives.' When the advice was
taken the deep trance was produced quite rapidly. This, of
course, is merely anecdotal evidence, but unfortunately very
little experimental work has been done on the form and con-
tent of suggestions made in inducing a trance, and conse-
quently nothing better than guesswork is possible on this
important point at the moment.

Nor, apparently, are the isolation, the quiet, and the
darkness important. Successful hypnoses have been carried
out under noisy conditions, in broad daylight, and even, as
is well known, on the stage in the presence of thousands of
people. Some hypnotists, in fact, claim that these conditions
are more favourable to the induction of a hypnosis than
those of absolute quiet and segregation. No experimental
work, again, has been done on this question, but it may be

34 Port One • Borderlands of Knowledge

surmised that different conditions suit different people, and
that whereas extraverts and hysterics may be more easily
hypnotized under conditions of noise, excitement, and while
in the limelight, introverted and anxious people might pre-
fer the quiet of the consulting-room. This, however, has by
no means been established as a fact.

Contrary to popular superstition, there is little difficulty
in awakening the subject once the experimenter decides to
end the trance. He usually suggests to the subject that when
he, the experimenter, counts up to ten, the subject will
awaken from his sleep, that he will forget everything that
has happened during the hypnotic trance, that he will feel
well and refreshed, and that he will feel all the better for his
experience. There are no records of any difficulties in awak-
ening subjects along these lines, and even if the hypnotist
should, for some reason, be unable to break off the hypnosis,
all that would be likely to happen would be for the subject
to fall into an ordinary sleep and wake after a few hours with
no evil after-effects.

Having induced a reasonably deep hypnotic trance in our
subject, what types of phenomena can be elicited? The first
and most obvious one, which, indeed, may be responsible in
large measure for all the others, is a tremendous increase in
the subject's suggestibility. He will take up any suggestion
the hypnotist puts forward and act on it to the best of his
ability. Suggest to him that he is a dog, and he will go down
on all fours and rush around the room barking and yelping.
Suggest to him that he is Hitler, and he will throw his arms
about and produce an impassioned harangue in an imita-
tion of the raucous tones of the Fuhrer! This tremendous in-
crease in suggestibility is often exploited on the stage to in-
duce people to do foolish and ridiculous acts. Such practices
are not to be encouraged because they go counter to the ideal
of human dignity and are not the kind of way in which
hypnosis ought to be used ; nevertheless, they must be men-
tioned because it is probably phenomena such as these
which are most familiar to people from vaudeville acts,
from reading the papers, and so forth.

Hypnosis and Suggestibility 35

It would not be true to say, however, that all suggestions
are accepted, even in the very deepest trance. This is par-
ticularly true when a suggestion is made which is contrary
to the ethical and moral conceptions held by the subject.
A well-known story may be quoted to illustrate this. Char-
cot, the great French neurologist, whose classes at one time
were attended by Freud, was lecturing on hypnosis and was
demonstrating the phenomena of the hypnotic trance on a
young girl of eighteen. When she had been hypnotized
deeply he was called away, and handed over the demonstra-
tion to one of his assistants. This young man, lacking the
seriousness of purpose so desirable in students of medicine,
even French ones, suggested to the young lady that she
should remove her clothes. She immediately awakened
from her trance, slapped his face, and flounced out of the
room, very much to his discomfiture.

Occasionally, failure to obey a suggestion is quite un-
accountable, and does not seem to be based on any kind of
ethical or moral scruples. In one of my early experiments I
was concerned with the ability of hypnotized people to judge
the passage of time. One subject in particular seemed to give
very accurate estimates of the number of seconds which had
passed between two signals given by the experimenter, and
in an attempt to make him improve on this performance it
was suggested to him that he was a clock. For some obscure
reason he did not take at all kindly to this idea and repeated,
in a sort of agitated way, that he was not a clock and that he
did not see how he could be a clock. To make the suggestion
more acceptable, he was then told that of course he was a
clock - couldn't he hear himself tick? This seemed to pacify
him for a little while, but he kept listening carefully, and
finally, in great agitation, denied that he could hear himself
tick. He grew so annoyed that it became necessary to aban-
don the idea and reassure him that he was not a clock at all.
As he had shown no objection to being a great variety of
other things more objectionable than clocks, this sudden re-
fusal was rather mysterious, and to this day its exact signi-
ficance escapes me. Generally, it is not at all unusual to find

36 Part One • Borderlands of Knowledge

the hypnotized person taking exception to certain sugges-
tions which, to the hypnotist, appear quite innocuous. Pos-
sibly, in these cases, there is some hidden significance pecu-
liar to the subject which would require an elaborate experi-
ment to unearth.

The next phenomenon frequently observed is known as
that of positive hallucinations. Here the subject will see,
hear, and feel the presence of objects which, objectively, are
not present at all. Tell him that his fiancee is sitting in the
chair opposite and he will greet her, go across and kiss her,
and generally behave as if his fiancee were actually there.
Tell him that a lion has just come in through the window and
he will show all signs of cringing fear, and may rush out of
the room in terror.

The converse of positive hallucinations are negative hallu-
cinations, which can also be easily induced. Here the subject
fails to see, feel, or hear objects and persons which are, in
fact, actually present. Suggest to him that he and the hypno-
tist are alone in the room and he will pay no attention to
other people and behave as if they were not there. Suggest
to him that he cannot feel any touch on his skin, or that he
cannot hear a certain sound and he will, in fact, behave as
if that were true. Positive and negative hallucinations of this
kind are relatively easy to produce in susceptible subjects,
but the criticism is often made that we may be dealing with
a simple desire on the part of the hypnotized person to
please the experimenter and that all these manifestations are
faked in some way. Alternatively, it is suggested that hypno-
tized people are in reality in the pay of the hypnotist and
merely pretend to go through the hoop in response to
favours received. This criticism may sound reasonable to
anyone who has not, in actual fact, seen the difference in be-
haviour between a person merely pretending to be hypno-
tized and a person in a trance. Nor does it account for
phenomena such as this. A stage hypnotist had hypnotized
a rather pompous, well-dressed young man on the stage, and
had got him to take off his trousers and to ride around the
stage on a broom-stick. The public were shrieking with

Hypnosis and Suggestibility 37

laughter at this sight, but when the hypnotist wakened the
subject from the trance, the latter took one look at his rather
undignified appearance, picked up the broom, and knocked
the hypnotist flat. It is difficult to consider this as part of the
act!

Probably more convincing is the fact that certain hallu-
cinations can be produced by hypnosis which give rise to
acts impossible to imitate in the normal state. Take a glass
full of soapy water and suggest to a hypnotized person that
it is bubbling champagne. He will drink it down with every
sign of enjoyment. This is a very difficult feat to encompass
in the normal state. The reader who doubts it might like to
carry out the experiment on himself! Similarly, it is possible
for a normal person to pretend to feel afraid of the non-exist-
ing lion who comes through the window, but he would find
it very difficult indeed to produce all the autonomic and
physiological signs of fear which are not under voluntary
control, and which nevertheless can be shown to be present
in the hypnotized person. In general it might be said that
while a certain amount of faking undoubtedly does occur in
stage performances, it would be very difficult, if not im-
possible, for such fakes not to be detected in the psychologi-
cal laboratory.

The unique nature of hypnotic phenomena becomes
even more apparent when we turn to another field which
has been extensively investigated. In connexion with nega-
tive hallucinations I mentioned the possibility of producing
anaesthesia, i.e. a failure to feel a touch applied to the skin
of the subject. It is similarly possible to produce a complete
insensitivity to pain, usually referred to as hypnotic anal-
gesia. For many years this phenomenon has been the subject
of doubt and derision, possibly because most of the pheno-
mena described so far might just have been capable of simu-
lation and faking. Too much is known, however, about
people's response to pain to leave one in any doubt that if
the phenomena described by hypnotists are, in fact, accur-
ate, then we are dealing in hypnosis with something quite
beyond the usual.

38 Part One • Borderlands of Knowledge

Let us start with a simple demonstration. The deeply
hypnotized subject is told that a needle will be pushed
through his hand and that he will not feel any pain what-
soever. He is also told that there will be no bleeding. The
hypnotist then pushes a needle through the subject's hand;
the latter does not even look at it and goes on talking as if
nothing had happened. There is no bleeding, or else very
little. Again the reader who believes that hypnotic pheno-
mena can be faked might like to try the experiment on him-
self!

The pushing of a needle through the hand is one thing ;
major operations are quite another. Yet there is no doubt
that in literally thousands of cases, major amputations have
been carried out under hypnosis without pain, and without the
usual accompaniment of shock, and other traumatic physio-
logical indices. Much of the credit for the introduction of hyp-
nosis into this field goes to Elliotson, a young physician at one
of London's major hospitals during the middle of the last cen-
tury, and to Esdaile, a physician working in India. An-
aesthetics had not been discovered then, and any kind of
operation, particularly a major one, was a very bloody
affair, in the true sense of that term. Elliotson's experiments
with hypnosis were not well received by his colleagues, who
resented his crusading vigour and did not like the odd and
unorthodox nature of the method. Esdaile perfected the
technique, and finally his claims were investigated by a
special commission, which, though incredulous at first, was
forced to report that he had in fact succeeded in carrying
out major operations without any evidence of pain or shock
on the part of the persons operated upon. This caused consider-
able and acrimonious controversy, but the discovery of an-
aesthetics around this time caused medical people to sink
back with thankfulness into lethargic uninterestedness in
hypnosis and similar oddities, and rest content with the more
physical type of anaesthetic, such as ether and chloroform,
which they felt they could understand more easily. Nowa-
days hypnosis is very little used in medical treatment for the
purpose of the suppression of pain, although it is superior in

Hypnosis and Suggestibility 39

many ways to the best available anaesthetics. Occasionally
promising results are reported in childbirth, and more re-
cently in connexion with the extraction of teeth. Hypno-
dontics, as this new method is somewhat oddly called, has
already produced a large number of reports of bloodless and
painless extractions of teeth, and a few years ago a public
exhibition was given in the United States in which, before a
large group of dentists, two upper bicuspids and one lower
bicuspid, all on the right side, were extracted without the use
of drugs or chemical anaesthetics. The periosteum was lifted
away and the three teeth extracted without the slightest in-
dication of pain, and without bleeding, while the patient
remained in a deep trance. Post-hypnotically there was no
sign of bleeding or recollection of pain.

Another phenomenon which appears spontaneously, and
has been remarked on from the earliest days of hypnosis, is
that of rapport. By this is meant a special relationship obtain-
ing between the hypnotist and his subject, such that the
latter takes orders, accepts suggestions, and so on, only from
the former and not from anyone else. The Freudian concept
of transference, i.e. the existence of a special relation between
patient and therapist, is in many ways a watered-down ver-
sion of the notion of rapport. It is quite likely, in fact, that the
underlying rationale of these two phenomena, when care-
fully examined, will be found to be somewhat similar. Al-
though psychoanalysts have always protested against this
notion, ' there seems to be little doubt that suggestion, al-
though not necessarily of a hypnotic kind, plays a very im-
portant part in their treatment.

Rapport, once it is established, can be transferred to other
people at the command of the hypnotist. He may tell the
hypnotized subject, 'This is Mr Smith. I want you to carry
out everything he tells you, just as you would carry out
everything that I tell you.' Such a command establishes
rapport between the subject and Mr Smith, and in this way
rapport can be handed on through a whole series of people.
In the absence of such a voluntary transfer, the hypnotized
subject would not be at all suggestible to anything Mr Smith,

40 Part One • Borderlands of Knowledge

or anybody else, might say to him ; his rapport is entirely with
the person who carried out the original hypnosis.

Post-hypnotic amnesia, or a complete forgetting of every-
thing that happened under hypnosis, is a very frequent con-
comitant of the hypnotic trance. It was first encountered in
the case of Victor, the young shepherd hypnotized by the
Marquis de Puysegur, whom we have already mentioned.
When he awoke from the hypnosis he had no recollection of
anything that happened during that period. Such amnesias
are common in deep states of trance, and do not require to be
suggested to the subject. They are apparently a spontaneous
outgrowth of the deep hypnotic trance. When the trance is
less deep it may be necessary for the hypnotist to suggest
that the subject should forget everything that happened;
such commands are usually obeyed without difficulty, ex-
cept in the very slightest stages of hypnosis, where they may
be ineffective.

There is thus a continuity between the last moment be-
fore the subject sinks into the hypnotic trance and the first
moment when he is awakened, with a complete amnesia for
everything that happened in between. This may be illus-
trated by an experiment carried out on a rather boastful and
arrogant young man who came to the laboratory loudly pro-
testing to everyone within earshot that he did not believe in
hypnosis, that he knew nobody could hypnotize him, and
that he would soon show the experimenters up for a bunch
of incompetent fools. He kept on talking in this fashion
while the hypnotic suggestions were repeated to him, which
were to the effect that he would fall into a deep sleep when
the experimenter knocked on the table with a reflex hammer
held in his hand. The young man was just saying, "... and
furthermore, I don't believe for a minute that anybody with
as strong a will-power ...", when the experimenter rapped
on the table. The subject's eyes closed immediately, he
stopped talking, and fell into a reasonably deep trance. For
slightly over two hours a series of experiments were con-
ducted with him, which showed him to be a very good sub-
ject indeed. At the end of that period the suggestion was

Hypnosis and Suggestibility 41

made to him that he would awake without remembering
anything about this hypnotic period. The moment the ex-
perimenter again rapped on the table with a reflex hammer
he continued talking, saying, *. . . as I can possibly be hypno-
tized'. He was quite incredulous when told of what had
happened and only an agitated reference to his watch led
him finally to believe that he had actually been hypnotized.

If after awakening the subject we hypnotize him again,
then his second hypnotic state is in contact through memory
with the first, but whatever happened between the two hyp-
notic states is not recollected. However, these amnesias can be
removed by suggestion. If it is suggested to the subject at the
end of the hypnotic trance that he will recall everything
that has happened during the trance, there is usually little
difficulty in making him conscious of and remember what
would normally be unconscious and forgotten.

One of the most striking ways in which post-hypnotic
amnesia is found to work is in the case of another hypnotic
phenomenon, which has excited considerable interest from
the very day of its discovery. This is the phenomenon of post-
hypnotic suggestion. If, under hypnosis, the subject is given
a suggestion to be carried out at a given time, or after re-
ceiving a certain signal, he will carry out this suggestion, al-
though he may at the time not be in a hypnotic trance at all,
but may have returned to the waking state. Here is a typical
case record to illustrate this. The subject is hypnotized and
told that he will be awakened after ten minutes. He is
further told that some time after this the hypnotist will blow
his nose three times. Upon the receipt of this signal, the sub-
ject will get up, go out into the hall, pick up the third um-
brella from the left on the rack, go back into the room, and
put up the umbrella there. After a little while the subject is
then awakened ; he talks animatedly, has forgotten all about
his experience, and when questioned as to whether he has
been asked to do anything, seems astonished at the idea, and
certainly cannot recall anything of the kind.

When the experimenter blows his nose three times, the
subject becomes vaguely restless and uneasy; finally, he gets

42 Part One • Borderlands of Knowledge

up, leaves the room, picks up the designated umbrella,
brings it back into the room and puts it up. When question-
ed as to the reasons for his actions, he cannot, of course,
give the true reason because he is unconscious of it. Instead
he will make up as good a reason as he can. Thus, he may
say, 'Well, you know the old superstition about putting up
umbrellas in a building. We were talking just now about
superstition and I wanted to show you that I was not super-
stitious myself.' Many of these rationalizations are quite re-
markable in their ingenuity, and intelligent people in partic-
ular can usually find a good reason for doing almost anything
that has been suggested to them under hypnosis. What is
more, they apparently believe their own rationalizations
implicitly. The tendency of human beings to rationalize
their actions and to believe in their own rationalizations
implicitly is, unfortunately, a phenomenon too widespread
and too well known to have escaped the notice of philo-
sophers and psychologists from the very beginning of inter-
est in human actions. What is important in this demonstra-
tion is the way in which it becomes possible to control the
situation in such a way that the true cause of a person's con-
duct is known to the experimenter, but not to the person
himself. Oddly enough, this very powerful method of in-
vestigating the process of rationalization has not been used
to any considerable extent for experimental purposes; it has
largely remained an amusing demonstration and an after-
dinner game.

There is no doubt about the great strength of post-
hypnotic suggestion and its capacity to produce action. An
illustration may serve to make this clear. The subject of the
experiment was a well-known psychologist, deeply inter-
ested in the phenomena of hypnosis, and himself an experi-
mentalist of considerable standing in this field. His
personality was very stable and strong, with no traces of
neurotic weakness. He expressed a desire to experience the
phenomena of hypnosis at first hand, and was accordingly
hypnotized, falling into a reasonably deep trance. In the
trance it was suggested to him that upon a prearranged signal

Hypnosis and Suggestibility 43

he would get up from his chair, walk across the room,
and sit down in another chair. He was awakened from the
hypnosis, and after half-an-hour or so the prearranged signal
was given. He became a little agitated, began to look across
the room at the other chair, and finally said, ' I feel a strong
tendency to go across the room and sit on that chair. I am
sure you have given me a post-hypnotic suggestion to this
effect. Well, I'm damned if I'll do it!'

He continued taking part in the conversation, but became
more and more distracted and monosyllabic, until finally he
jumped out of his chair, crossed over, sat down in the de-
signated chair, and exclaimed, 'I couldn't stand it any
longer ! '

What apparently happens is that the post-hypnotic sug-
gestion sets up an encapsulated action tendency in the mind
which is relatively independent of voluntary control, and
powerfully demands action before it can be reintegrated
with the remainder of the subject's mind. In this it very
much resembles in miniature the kind of complex so often
found in neurotic and otherwise emotionally unstable
patients. The cause of this action tendency is unknown to
the subject, and even where it is guessed, as in the case of the
psychologist just mentioned, this knowledge does not seem
capable of counteracting the determining influence of this
small 'complex'. When it is remembered that in the
particular case just mentioned this single suggestion
triumphed over the strength and will-power of a well-
integrated, strong-willed, competent person, who, in fact,
had guessed what was happening, it will be realized that
hypnosis and hypnotic suggestions are no playthings, but
carry with them an almost frightening degree of strength
and importance.

In view of this, it is small wonder that attempts have been
made to use post-hypnotic suggestion as a curative agent.
This idea was particularly obvious when it became known
that post-hypnotic suggestion could last for very long
periods of time. Responsible investigators have claimed that
post-hypnotic suggestions have been carried out as much as

44 Port One • Borderlands of Knowledge

five years after they were originally given, and periods of
several months to a year have been vouched for by a num-
ber of trustworthy experimentalists. Yet, in spite of this, re-
sults of curative post-hypnotic suggestion have not usually
been very positive. Supposing you wish to give up the habit
of drugging yourself with alcohol or tobacco. It is possible to
give a suggestion to the effect that alcohol will make you
sick and that tobacco will taste like aloes. This suggestion
certainly works in the hypnotic trance, and even when given
as a post-hypnotic suggestion it works for a day or two
afterwards, but gradually the strength of the suggestion
seems to wane, until within a week or so nothing is left and
the subject returns to his original addiction. It is, of course,
possible to re-hypnotize him every few days and give post-
hypnotic suggestions again and again, but this course is not
favoured because many people fear that hypnosis itself will
become an addiction and possibly a worse and more expen-
sive one than either alcohol or tobacco. Consequently, at the
present time, little use is being made of post-hypnotic sug-
gestion, although it has always seemed to me that if the
method could be experimented with in a more whole-
hearted and intelligent manner than has been done so far
much more positive results might be obtained.

Another phenomenon often observed in hypnosis is the
transcendence of normal working capacity. Many absurdly
exaggerated claims have been made in this connexion, par-
ticularly by some of the earlier writers who were so much
impressed with what people could do under hypnosis that
they did not bother to find out whether much the same
could not be done in the normal state also ! However, a ker-
nel of truth does seem to be contained in the view that under
hypnosis certain activities are performed more quickly or
more accurately than in the waking state. In carrying
out such simple tasks as placing dots as rapidly as possible
into small squares, multiplying numbers, crossing out
certain letters in a given list, adding numbers, counting in
threes, tapping as fast as possible, sorting cards into packs,
placing rings on a pole, and so forth, it has been found that

Hypnosis and Suggestibility 45

improvements varying from 30 to 60 per cent can be pro-
duced through hypnotic suggestion. A rather smaller amount
can be produced by post-hypnotic suggestion, percentages
ranging from 20 to 40 per cent approximately.

When a study is made of all the reports on properly
planned and executed experiments in this field, it becomes
clear that the amount of improvement shown in a given
task is not invariable, but appears to be a function of the
complexity of the task. As a general rule it may be said that
the simpler the task the greater the improvement, and, conversely,
the more complex the task the less improvement will there be. In very
complex activities, such as doing an intelligence test, there
may not only be no improvement at all, but a slight drop in
performance may occur.

It is difficult to know to what causal factors one may
ascribe improvements of this type. One hypothesis, which
has a certain amount of experimental backing, is that they
are related to a decrease in fatigue, which is a well-known
concomitant of hypnotic suggestion. This is one of the most
noticeable features of hypnosis, and one which is recognized
by all workers in the field. The report of an actual experi-
ment may illustrate the difference between the normal and
the hypnotic state. The subject is asked to pull a dynamo-
meter in rhythm with a metronome. A dynamometer is an
instrument for measuring strength of grip. The subject pulls
the handle as strongly as he can; this handle is connected to
a steel spring, and the pressure exerted by him against the
spring is recorded on a dial. Pulling a dynamometer at fre-
quent intervals is a very tiring business and scores dropped
off rapidly. Figure 1 shows the amount of pull exerted by a
subject in the normal state (N) andalsowhen hypnotized (H).
Under hypnosis it was suggested to him that he was the
strongest man in the world and that he would show no
fatigue at all.

It will be seen that the suggestion of increased strength
produced practically no effect, as the starting points of the
two curves are very close together. The effect of the sugges-
tion of 'no fatigue', however, can be seen from the fifth trial

46

Part One • Borderlands of Knowledge

onwards. While the curve depicting the strength of pull in
the normal state drops until, after the twelfth trial, the sub-
ject is incapable of going on at all, under hypnosis it remains
at the same level after the fifth trial and the subject goes on
for quite a long time after the twelfth trial. This result is

4 5 6 7 8 9 10 II 12

Number of contractions

Figure 1 : Weight in grams pulled by subject in the
normal (N) and in the hypnotized (H) state

quite typical of many others achieved in this field, and as a
good number of the tasks on which hypnotized people show
superiority over the normal state involve muscular contrac-
tions of one kind or another, the possibility cannot be ruled
out that improvement is largely due to absence of fatigue in
the hypnotized condition.

Another hypothesis which has to be considered relates to
the possibility that the improvement in performance is really

Hypnosis and Suggestibility 47

due to changes in the emotional state of the subject. Many
people, when subjected to psychological tests, show a certain
amount of anxiety and react emotionally to the whole test-
ing situation. It will be shown in a later chapter that emo-
tion tends to interfere with fine muscular adjustment, and,
consequently, it might be thought that the more even emo-
tional tone induced as part of the hypnotic procedure, and
the dispersal of anxiety through suggestion, which also forms
part of the hypnotist's method, will improve performance
through a reduction in the amount of interference by
emotion. In many cases this may be a concomitant, but it is
unlikely to be the whole explanation. One or two reports
have indicated that, under certain circumstances, the re-
moval of anxiety and other disturbing emotions, through
hypnosis, may even improve performance on tests of in-
telligence. It should, however, be emphasized that this ap-
pears to be true only in cases of extreme emotional instabil-
ity, and that for normal people there is little likelihood of
their obtaining an improved I.Q,. through hypnotic
suggestion !

A third hypothesis relates to the question of motivation.
In the normal state many subjects may not be very strongly
motivated to carry out the simple tasks required of them by
the experimenter. Hypnosis may alter the degree of motiva-
tion and, in consequence, may produce improvements in
output. All these theories may be in part true; while the fact
of improvement in simple motor functions under hypnosis is
undisputed, its causes are not known for certain.

Another phenomenon of hypnosis which has attracted
probably more attention than almost any other, because of
its therapeutic promise, is the improvement of memory un-
der hypnosis. It is claimed that under hypnosis a person can
remember things which in the normal state he would be
quite unable to recall. In fact, it is suggested that under
hypnosis a person can be 'regressed' to an earlier age and
that in this condition he will experience again the events
which were happening at that time, and the emotions which
they evoked in him. This claim has led to a considerable

48 Part One • Borderlands of Knowledge

amount of critical discussion, particularly because some of
the advocates of hypnotic regression have gone to rather ex-
treme lengths in their claims. Thus, not only has it been said
that under hypnosis people can be regressed to the moment
of birth, but claims have even been made that memory of an
intra-uterine existence has been achieved. Absurd claims of
this nature have not helped to make the subject popular
with critical scientists, but there is some evidence available
which is of considerable interest and importance.

First of all, let us look, from a purely descriptive point of
view, at the sort of thing that happens when a subject is
hypnotized and regressed to an early age. The subject
usually employs language such as might be expected of a
person of the regressed age. His voice may become child-
like, or he may retain a normal tone but use only simple
words and phrases. His general behaviour will tend to con-
form to the suggested age. If regressed to the age of five, say,
he will play with toys and react with protests and tears if the
toys are taken away. Drawings will be child-like and corre-
spond roughly to the level of a child of the regressed age.
The handwriting changes and frequently becomes very like
the actual specimens of the subject's childhood writing. In
one case it was found that when a twenty-year-old girl was
regressed to various ages she changed the chalk to her left
hand at the six-year level ; she had started writing with the
left hand, but had been forced to change over at the age of
six.

In another case, a male subject, thirty years old, was
seated in a chair arranged in such a way that the release of
a latch would cause it to fall back into a horizontal position.
(This is a favourite torturing device much used by psycho-
logists to stimulate emotional reactions!) When the man
was regressed to a level of about one year of age, the latch
was suddenly released so that the chair fell back. An adult,
or an older child, would quite involuntarily extend both
arms and legs in an effort to maintain balance. The re-
gressed subject, however, screamed in fright, but made no
movement of the limbs and fell backward with the chair.

Hypnosis and Suggestibility 49

Another reaction, quite unexpected and embarrassing, both
to the hypnotist and to the subject, was an accompanying
urination which soaked the man's trousers! It is unlikely
that such behaviour is simply due to play-acting.

Other types of investigation have been concerned with
the behaviour of regressed subjects in tests of intelligence,
and in various types of achievement tests. It is usually found
that when regressed to a certain level, people tend to behave
on tests of this type in a manner roughly appropriate to the
given age. Such reactions, of course, could easily be faked,
but it has also been shown that when, for instance, the eye
movements of subjects are photographed, a considerable
lack of ocular co-ordination and stability is found when re-
gression to a relatively young age occurs. Such physio-
logical phenomena are characteristic of young children and
are difficult, if not impossible, to produce voluntarily.

A similar effect has been reported in the case of a subject
whose eyesight had been defective since early childhood,
and who had begun wearing glasses at the age of twelve.
Regression to the age of seven produced a definite improve-
ment in both near and far vision, as measured by an opto-
metrist. Even more impressive is another case of a subject
who had had a colloid cyst removed from the floor of the
third ventricle. Prior to this removal, the subject had been
suffering from blindness in the left half of the right eye. After
the operation, vision had become normal, but when the sub-
ject was regressed to a time shortly before the operation the
visual defect again reappeared during the regression.

A rather different type of proof is of a purely neuro-
physiological nature and concerns the Babinski reflex. In a
normal adult, stroking the sole of the foot causes the great
toe to turn down. In infants up to approximately seven
months of age, however, the reflex response to stroking the
sole of the foot is dorsiflection, or the turning up of the toe.
If regression were a real phenomenon, then we would ex-
pect that, on regressing adults to the age of five or six
months, the Babinski reflex, i.e. dorsiflection, should occur.
This has actually been found to be so in subjects who could

50 Part One • Borderlands of Knowledge

not be expected to have known these rather esoteric details
of neurophysiological development.

In discussing some of the evidence, I have not mentioned
the many studies in which patients have been asked to re-
member certain events which happened at a relatively early
stage of their lives, and where later checks revealed that
these had actually happened. This type of work is too much
open to falsification and all sorts of uncontrolled influences
to be of much value. The alleged events which the hypno-
tized person experiences under regression may have been
discussed with him by other people long after they had in
fact happened ; the memories of the witnesses may them-
selves be affected by the story told by the subject who is
being regressed; also, certain confirmative details may be
elaborated by the experimenter to the exclusion of items
that had not fitted in. Quite recently, however, this unsatis-
factory type of evidence has been transformed into a scien-
tifically useful and quite decisive method of experimenta-
tion. The idea is so simple, and so lacking in any technical
complexity, that the reader might like to exercise his own
ingenuity in deciding how he would solve this particular
problem. The patient under hypnotic regression tells of cer-
tain events and memories of things that happened many
years ago. How can the truthfulness of these memories be
checked without relying on subjective memories of other
people? The answer to this problem is a very simple one, but
any reader who can think out the right answer has shown
more scientific inventiveness than hundreds of academic and
medical hypnotists who have repeated the same type of in-
valid and useless investigation time and time again.

The solution essentially consists in finding an objective
fact which at the time to which the subject is regressed
would be well known to him, but which during the course of
his life he would certainly forget. Facts of this kind are the
day of the week on which his fourth, or eighth, or tenth
birthday fell, or the day of the week on which a certain
Christmas celebration fell, and so on. The procedure of test-
ing is a very simple one. The subject is asked on what day of

Hypnosis and Suggestibility 5 1

the week, say, his sixth birthday fell. Practically no one suc-
ceeds in correctly remembering this far-off event, which
took place twenty or more years ago. He is then hypnotized
and gradually regressed to this particular day. Now birth-
days are of the very greatest importance to children, and
they know perfectly well at that time what day of the week
their birthday is. Consequently, having been regressed to
the day, the subject is simply asked what day of the week it
is. Correct answers have been obtained for 93 per cent of
subjects, regressed to the age often; 82 per cent of subjects
regressed to the age of seven; and 69 per cent of those
regressed to the age of four.

Experiments such as those described in some detail above
leave little doubt that there is a substantial amount of truth
in the hypothesis that age regression does, in fact, take place,
and that memories can be recovered which most people
would think had been completely lost. This fact may be
used for psychotherapeutic purposes, and an illustrative
case may show how this can be done. The patient is a mar-
ried woman of forty-two, intelligent and well-informed. For
many years this patient, Mrs Smith, has suffered from re-
current asthmatic attacks. Her work necessitated her enter-
ing various hospitals (she was a psychiatric social worker),
and in such a situation she always experienced a very strong
fear reaction. Other unreasonable fear reactions were pro-
duced by the sight of a pair of hairy arms on a man. Knives
also produced considerable fear in her, and she sometimes
experienced nightmares.

During a self-induced trance one day, she was regressed
to an early age, when she experienced a previously com-
pletely forgotten incident with unusual clarity. She seemed
to be lying on a table under brilliant lights. A man was
standing beside her holding a small knife. A vague, threat-
ening object was descending from above her head, and
settled down over her face. She was terror-stricken and
tried to rise, but two hairy arms grabbed her and roughly
forced her back. She continued to struggle, but was
violently shaken and slapped repeatedly by someone.

52 Part One • Borderlands of Knowledge

Finally, the object came down over her face and smothered
her.

On inquiry, it was found that at the age of sixteen months
a mastoidectomy had been performed on her and that she
had been very sick afterwards with complications caused by
severe shock. Two of the nurses at the hospital had told her
mother about the brutality displayed towards the child by
the anaesthetist, and they had resigned in protest. For some
time afterwards the child experienced nightmares and had
been emotionally disturbed. It was following this operation
that Mrs Smith's first attacks of asthma had occurred.

The result of this 'Mesmeric crisis' or 'Freudian abreac-
tion ' was that Mrs Smith has been entirely free of asthma
ever since ; she lost her fear of hairy arms, and the phobias
of knives and hospitals have completely disappeared.

Two further points may be mentioned next in connexion
with our discussion of the phenomena characterizing
hypnosis, although they are only indirectly related to this
topic. One is the question of whether people can be induced
to commit criminal acts under hypnosis ; the other how many
people are capable of being hypnotized. Both are questions
probably more frequently asked than any others in connexion
with hypnosis. Taking the question of the production of
criminal activity in hypnotized persons first, it may be said
that until fairly recently the more sober writers tended to
discountenance this possibility. They tended to quote the
case of Charcot's young assistant, who failed to induce the
young hypnotized girl to take off her clothes, and to infer
that, quite generally, a suggestion urging a person to act in
ways which were very much counter to his moral and ethical
ideas would not be carried out, but would merely lead to his
awakening. There are, indeed, many observations of this
kind to be found in the experimental literature, and it may
be said with a reasonable degree of confidence that in many
cases an explicit suggestion to do something unethical or immoral
will not be carried out by the subject.

More recently, however, a number of experiments have
been conducted to show, first, that this conclusion is not

Hypnosis and Suggestibility 53

universally true, and, secondly, that the whole framework
of the type of experiment on which it is based is much too
narrow. One example may suffice to show the kind of experi-
mentation involved. The experimenter demonstrated the
power of nitric acid to the subject by throwing a penny into
it. The penny, of course, was completely disintegrated and the
subject began to realize the tremendous destructive power
of nitric acid. While the subject's view of the bowl of acid was
cut off by the experimenter, an assistant substituted for it a
like-sized bowl of methylene-blue water, continuously kept
boiling by the presence in it of miniscule droplets of barium
peroxide.

The hypnotized subject was then ordered to throw the dish
of nitric acid (in actual fact, of course, innocuous water)
over the assistant who was present in the same room. Under
these conditions it was possible to induce, under hypnosis,
various subjects to throw what they considered to be an ex-
tremely dangerous acid into the face of a human being. It
might be argued that perhaps they had noticed the differ-
ence between the acid and the water. Actually, in this par-
ticular experiment, the person in charge made what he calls
ca most regrettable mistake in technique' by forgetting to
change the nitric acid to the innocuous dish of water, so that
in one case the assistant had real nitric acid thrown over
him. (Because of the promptness of remedial measures, no
scars were left on the face of the assistant.) This touch of
realism was not intended, but it does illustrate that even the
experimenter and the assistant could not tell the true acid
from the fake one.

Another experiment reports how a post-hypnotic sugges-
tion was implanted in a soldier assigned to military duty.
The post-hypnotic suggestion was carried out and the sol-
dier deserted his duty, an anti-social act which would un-
doubtedly have resulted in severe punishment if the cir-
cumstances had not been known. It will be seen, therefore,
that certain anti-social activities can be produced by
hypnosis, although they might not seem to the reader to be
of a very serious nature. The difficulty, of course, is that if

54 Part One • Borderlands of Knowledge

the misdemeanour is of a serious nature, then the execution
of the act would quite rightly result in punishment by the
courts for the hypnotist, and perhaps also for the person
hypnotized.

However, much more important than these demonstra-
tions that direct suggestion can produce anti-social activities
is a rather different line of work which is based on an im-
portant theoretical consideration. It is maintained thai, by
making a direct suggestion to the subject, the hypnotist does
not, in fact, make the best use of the known phenomena of
hypnosis. We may consider as an example the case, already
several times mentioned, of the young girl who refused to
take off her clothes. Speculating on her state of mind, we
might argue that she would be exposed to a considerable
conflict. On the one hand, there is the powerful suggestion
to undress ; on the other hand, there is the obvious presence
of a group of young men, resulting in a conflict between the
suggestion and the ethical and moral ideas inculcated in her
through a long process of education.

Now, if the hypnotist were serious in his attempt he
would, of course, set about it in quite a different way. He
would, by means of suggestion, attempt to remove the con-
flict in the girl's mind. He would do this first of all by induc-
ing negative hallucinations in her to the effect that she was
alone in the room, or possibly that only she and a girl friend
were in the room, the experimenter casting himself in the
role of the girl friend. He would then go on to use positive
hallucinations to suggest that they were both in the girl's
bedroom, spending quite a while implanting this suggestion
in great detail in the subject's mind, and painting very
vividly the existence and position of different articles of
furniture.

The next suggestion would be, perhaps, to the effect that
it was getting late, that the light was fading, that they had
to get up early next morning, and that they ought to be
going to bed. This kind of suggestion, once accepted, could
then rapidly lead to the suggestion that they should now
take off their clothes and go to bed. There is little doubt

Hypnosis and Suggestibility 55

that, under these conditions, there would be no difficulty at
all in producing the desired result. (The experiment has not,
to my knowledge, been done, but evidence will be quoted
below to leave little doubt on this score.)

Essentially, the technique suggested for achieving anti-
social results is not so very different from that used by many
people in ordinary life. Let us take the case of Sir Pumper-
dinck Flannel Flannel, the moustachioed cad in a Victorian
melodrama. Bent on the seduction of the innocent maiden,
what does he do? He obviously does not go up to her and
say, 'To bed with you, my proud beauty'. What he does in-
stead is the well-known routine of trying to create the im-
pression in the girl's mind of a situation which, in fact, does
not exist. He pretends to eternal love; he pretends to desire
to marry her; he makes a large number of vows designed to
confuse the issue and make his desires appear in a different
light. The result, as is shown by the published statistics on
illegitimacy, is often quite favourable, in spite of the absence
of hypnosis. When hypnosis is added to the picture there is
little doubt about the much greater efficacy of this type of
method.

From speculation to experimentation. Here is an experi-
ment on a private soldier, about twenty years of age, who
had a very good Army record. He was placed in a trance
while several senior Army men were present. A lieutenant-
colonel was placed directly in front of the subject, about ten
feet away from him, and the subject was then put in a trance
and the following suggestions made to him : * In a minute you
will slowly open your eyes. In front of you you will see a
dirty Jap soldier. He has a bayonet and is going to kill you
unless you kill him first. You will have to strangle him with
your bare hands.'

The subject opened his eyes, began to creep forward very
slowly, and finally, in a flying tackle, brought the lieutenant-
colonel down on the floor, knocked him against the wall,
and began strangling him with both his hands. It needed
three people to pull him off and break his grip, and he could
not be pacified until the experimenter had put him back

56 Part One • Borderlands of Knowledge

into a deep, quiet sleep. The attacked man reported that
there was no make-believe about the attack and that he
might have been killed or injured had assistance not been
immediately available. When it is remembered that to at-
tack a commissioned officer is a very serious offence in the
Army, it will readily be seen that a skilled hypnotist, by mis-
representing the situation to the subject, could easily pro-
voke an anti-social act of very great severity indeed. When
we add to these considerations the fact that the subject may
be hypnotized against his will, we can see that the easy belief
that not much harm can be done under hypnosis, because of
the safeguards latent in the individual's ethical and moral
code, is dangerously complacent. Much work will need to be
done before we know the possible limitations of hypnotic
control over other people, but the dangers inherent in the
anti-social use of hypnosis should certainly not be mini-
mized.

The seriousness of these dangers is probably related to the
proportion of people in whom a deep hypnotic trance can be
induced. This proportion, unfortunately, is not known; in-
deed, in principle the problem as stated is an insoluble one.
We cannot, in the nature of things, say what proportion of
the population is tall, or intelligent, or fat, because these are
graded qualities and there is no point at which we can say
anybody beyond this point is tall, or intelligent, or fat, and
anybody who falls short of it is the opposite. Similarly,
hypnotic susceptibility forms a graded continuum, different
people being susceptible to different degrees, and there is no
one point where one could say that anyone beyond it could
be called hypnotizable. This continuous nature of hypnotic
suggestibility is demonstrated fairly clearly in a number of
'hypnotic scales', which have been drawn up to measure the
depth of trance. These scales make use of a number of sug-
gestions of increasing difficulty, recording in each case the
success or failure of the hypnotized person with respect to the
suggestion employed. The importance of a particular
phenomenon is assessed by taking into account its rarity, i.e.
the difficulty of eliciting it in a large group of people who are

Hypnosis and Suggestibility

57

being hypnotized. Table i below illustrates the construction
of one such scale. The suggestions given are printed on the
left ; the number of points given to each item indicate the
number of instances in which that particular suggestion was
successfully employed. Thus, a large number of people
accepted the suggestion that their eyes were feeling tired,
that they were completely relaxed and incapable of activity.
A glove anaesthesia, i.e. the incapacity to feel a touch over
an area of skin covered by a glove, was distinctly more diffi-
cult to induce. The illusions of bell-ringing and of leg- move-
ment were even more rare. The illusion of an electric bulb
lighting up, a post-hypnotic suggestion to the effect that the
subject should shake and open a box, and spontaneous
amnesia to the whole process were rarest of all. Each in-
dividual could be given a score on this scale according to

TABLE I

Eysenck-Furneaux Scale

Items Suggested

Points

Eyes tired

76

Complete relaxation

76

Feels incapable of activity

65

Arm falls irresistibly

63

Eyelids heavy

61

Impossible to raise arm

59

Feels miles away

54

Feels pleasant warmth

53

Glove anaesthesia

46

Eyes closing

45

Both arms stiff and rigid

36

Impossible to raise arm (eyes open)

24

Illusion of bell-ringing

22

Complete catalepsy

21

Cannot hear buzzer

19

Illusion of leg movement

18

Increase of Body Sway (post-hypnotic)

18

Line length suggestibility (post-hypnotic)

16

Amnesia (spontaneous)

15

Shake and open box (post-hypnotic)

13

Illusion of electric bulb lighting up

12

58 Part One • Borderlands of Knowledge

the degree of susceptibility shown by him, and these scores
could be shown to form a regular progression or continuum
from high to low.

Some authors, instead of using a continuous scale, have
preferred to use categories ; thus, one author uses the follow-
ing five categories:

(1) Insusceptible : total lack of response to suggestion.

(2) Hypnoidal: relaxation, fluttering of the eyelids, closing
of the eyes ; complete physical relaxation.

(3) Light Trance: Catalepsy of the eyes; limb catalepsies;
rigid catalepsies; glove anaesthesia.

(4) Medium Trance: Partial amnesia; post-hypnotic an-
aesthesia; personality changes; simple post-hypnotic
suggestions.

(5) Deep Trance: Ability to open the eyes without affect-
ing the trance; bizarre post-hypnotic suggestions;
post-hypnotic hallucinations; positive and negative
auditory hallucinations; systematized post-hypnotic
amnesias.

It is possible, from the writings of many different experi-
menters, to give rough percentage figures of people falling
into these various categories. About 15 per cent are found
to be insusceptible; about 40 per cent are found to fall into
a hypnoidal or light trance; some 25 per cent show pheno-
mena characteristic of a medium trance; and some 20 per
cent show phenomena characteristic of a deep trance.

These values are representative but, unfortunately, their
meaning is not too clear. There are several reasons for this.
In the first place, hypnotic behaviour can be learned and
practised very much in the same way that other types of be-
haviour are learned and practised. A person may start out
as a rather indifferent subject, but after a certain amount of
practice he may develop deeper and deeper stages of
hypnosis. If he is classified according to his reactions right at
the beginning, he would be considered insusceptible or
hypnoidal only ; if classified after several hours of practice,
he might be classified as capable of a deep trance.

Hypnosis and Suggestibility 59

Another difficulty which arises is connected with different
methods used by hypnotists; the different capabilities of
hypnotists ; and the different degrees of practice which they
have had. Thus, some people are considerably better than
others in inducing a deep hypnotic trance, and conse-
quently the per cent figures obtained by them would be
quite different from those obtained by an indifferent hypno-
tist. Nor can it be said that the hypnotist is necessarily
'good' or 'bad' in absolute terms. A particular hypnotist
may succeed with Mr Smith and fail with Mr Brown, while
another hypnotist may succeed with Mr Brown and fail with
Mr Smith. There are complexities in these personal relation-
ships which very much complicate the picture.

Another complicating feature is the particular technique
employed by a given hypnotist on a given occasion, and the
length of time he is willing to continue trying. One of the
best-known modern hypnotists reports that one of his most
capable subjects required less than thirty seconds to develop
his first profound trance, while a second, equally competent
subject, required 300 hours of systematic labour before a
trance could be induced. Few experimenters have spent as
much time as this on their less successful subjects; it is quite
possible that if they had done so most of these would have
been put into a higher category. One report will illustrate
the chance element which makes most of the quoted figures
somewhat meaningless. A very determined hypnotist spent
several hours on a given subject without inducing any kind
of hypnotic reaction whatsoever. Finally, at the end of a
three-hour session he completely lost his temper and shouted
at the subject, 'For — 'ssake go to sleep you !' The sub-
ject immediately fell into a deep trance and was an ex-
emplary subject ever after.

To these many difficulties are added the frequent changes
in motivation which undoubtedly play a role in determining
a person's reaction to these hypnotic situations. It will be
realized that the simple question regarding the proportion
of people who were hypnotizable in the population does not
admit of any kind of straightforward answer. There are so

60 Part One • Borderlands of Knowledge

many qualifying conditions, and so little experimental work
to clarify their influence that not even a guess would seem
justified. What can be said, however, is that in the hands of
a competent hypnotist who is willing to spend at least four
hours on each person, and under conditions which make the
subject reasonably co-operative rather than hostile, hypno-
tic phenomena can be induced in something like 85 per cent
of the population. It is quite possible that this figure should
be higher still, and, indeed, it might easily go up to 100 per
cent. Thus, hypnosis is not a rare and isolated phenomenon,
but one which is relevant to the great majority of people,
and possibly to everyone.

We have now covered the main events characterizing the
hypnotic state; we may next turn to some of the theories ad-
vanced by various people to account for these phenomena.
Mesmer's theory of 'animal magnetism' was taken over by
some of his followers, who became interested in hypnotic
phenomena. It should hardly be necessary to argue against
this ancient doctrine, but we may note some of the people
who were instrumental in proposing alternative theories.
Mainly involved in this liberation from the absurdities of
Mesmer's views were James Braid, an English physician,
and a group of French investigators. Braid, who became in-
terested in hypnotic phenomena through seeing a public
demonstration of them, was at first sceptical, but later became
convinced of the reality of the phenomena displayed. He con-
ducted a series of researches, and these very soon caused him
to reject the view that hypnotic phenomena were due to a
fluid passing from the body of the hypnotist to that of the
subject.

Braid also coined the word 'hypnotism', introduced the
method of hypnotism described early in this chapter, and
still most widely used, and also utilized the trance for pain-
less surgical operations. In addition, he was one of the first
to realize that suggestion played the major role in hypnosis,
a conclusion which had already been adumbrated by the
Royal Commission, whose report on Mesmer's work has
been quoted earlier on.

Hypnosis and Suggestibility 61

While these things are to Braid's credit, it should be noted
that much of his experimental work was thoroughly bad and
gave rise to beliefs hardly less absurd than those of Mesmer.
Braid was a firm believer in phrenology, that curious view
which teaches that human beings are possessed of certain
faculties; that these faculties are located in certain parts of
the brain ; that the degree of development of a given faculty
is indicated by the physical development of the correspond-
ing part of the brain; and that this physical development is
mirrored in the contours of the skull. Phrenologists accord-
ingly feel the bumps of the skull in order to arrive at their
analysis of a person's temperament and abilities. Braid con-
tributed to this view by teaching that if pressure was applied
to the various areas of the heads of hypnotized persons, then
behaviour characteristic of the corresponding phrenological
faculty would be elicited. Thus, he writes that when he
pressed the bump, or 'organ', of veneration on the head of
the patient, 'an altered expression of countenance took
place, and a movement of arms and hands, which later be-
came clasped in addition, and the patient . . . arose from the
seat and knelt down as if engaged in prayer'. It is only fair
to say that Braid gradually improved the standards of his
experimentation and that he finally gave up his belief in
phrenology. The fact, however, that such an able and com-
pletely honest person as Braid committed these very ele-
mentary errors suggests both the difficulty of experimental
work in this field, and also the very undeveloped state of
psychological experimentation. There will be occasion later
on to return to this point.

In France a comparable development took place, largely
due to the work of a humble physician in Nancy, who prac-
tised hypnotism on the poor peasants who came to his house.
This man, Liebeault, has rightly become famous, not only
for his contribution to the study of hypnosis, but also be-
cause of his unselfish generosity. He refused to accept fees
for his hypnotic treatments, and the lovable nature of the
man may be seen from a description of his clinic given by
Bramwell, a well-known British historian of hypnosis :

62 Part One • Borderlands of Knowledge

His clinique, invariably thronged, was held in two rooms in the
corner of this garden. . . . The patients told to go to sleep appar-
ently fell at once into a quiet slumber, then received their dose of
curative suggestions, and when told to awake, either walked quietly
away or sat for a little to chat with their friends, the whole process
rarely lasting longer than ten minutes. . . . No drugs were given, and
Liebeault took special pains to explain to his patients that he
neither exercised nor possessed any mysterious powers, and that all
he did was simple and capable of scientific explanation. ... A little
girl, about five years old, dressed shabbily, but evidently in her
best, with a crown of paper laurel leaves on her head and carrying
a little book in her hand, toddled into the sanctum, fearlessly inter-
rupted the doctor in the midst of his work by pulling his coat, and
said, 'You promised me a penny if I got a prize'. This, accom-
panied by kindly words, was smilingly given, incitement to work
having been evoked in a pleasing, if not scientific way. Two little
girls about six or seven years of age, no doubt brought in the first
instance by friends, walked in and sat down on a sofa behind the
doctor. He, stopping for a moment in his work, made a pass in the
direction of one of them, and said, 'Sleep, my little kitten', re-
peated the same for the other, and in an instant they were both
asleep. He rapidly gave them their dose of suggestion and then
evidently forgot all about them. In about twenty minutes one
awoke, and wishing to go, essayed by shaking and pulling to
awaken her companion - her amused expression of face, when she
failed to do so, being very comic. In about five minutes more the
second one awoke, and, hand in hand, they trotted laughingly
away.

After many years of hard work, Liebeault put his views
into a book, the main teachings of which were similar to
those of Braid. However, as was perhaps inevitable in one so
self-effacing, only one copy of this book was sold, and it was
not until twenty years later, when Bernheim, a Professor in
the medical school at Nancy, became acquainted with Lie-
beault, that the attention of the medical profession was
directed towards Liebeault's great successes in treating
various types of disorders by means of hypnosis, and to his
theories attributing these effects to suggestion.

It might have been thought that when serious scientists
had exorcised the devil of 'animal magnetism' in this

Hypnosis and Suggestibility 63

thoroughgoing fashion, and when so much experimental
work was available to show the influences that were really
operating, a period of patient, quiet research would follow,
untroubled by arguments about the ancient Mesmeric doc-
trines. However, this was not to be, and the person respon-
sible for the revival of the doctrine of animal magnetism was
a well-known French anatomist and neurologist named
Charcot. It is one of the tragedies of science that Charcot,
who was an extremely able experimenter in physiology, is
known to posterity largely because of the egregious errors
committed in his experimental studies of hypnosis. The
story of how this came about is a fascinating one, and also
instructive, as it shows clearly that competence in physio-
logical experimentation does not, by any means, guarantee
competence in psychological work, and may in fact blind
the person in question to sources of error of a purely psycho-
logical nature.

Charcot was very much afraid of being deceived by his
subjects, and determined that his experiments should be as
rigorous and ultra-scientific as his physiological work. Con-
sequently, he sought diligently in the behaviour of his sub-
jects for signs of hypnotic trance behaviour which could not
be simulated and which were of a completely objective
nature. He concluded from his studies that hypnotism
showed three definite stages: lethargy, catalepsy, and
somnambulism. The lethargic stage, which was induced by
closing the subject's eyes, was characterized by an inability
on the part of the subject to hear or to speak. Also contrac-
tures of a specified nature resulted when certain nerves were
pressed. The cataleptic stage was produced by opening the
subject's eyes while he was in the lethargic stage. Now limbs
would remain in any position in which they were placed by
the experimenter, and the subject was still unable to hear or
speak. The somnambulistic stage could then be induced by
applying friction to the top of the head, and it is this
somnambulistic condition which appeared similar to what
we would now call an ordinary hypnotic trance.

Another phenomenon much stressed by Charcot was that

64 Part One • Borderlands of Knowledge

of transference. Sometimes he found that the contractures,
catalepsies, and so on, would appear on only one side of the
body. If now a large magnet was brought close to the limbs
in question, the symptoms could be displaced at once to the
other side of the body.

Bernheim and Liebeault reported that these alleged stages
of hypnosis were never found in their own work unless the
subjects had been taught to expect to go through them. In other
words, Charcot's allegedly 'objective' signs of hypnosis had,
in fact, been suggested by him in some form or other to his
subjects and had no necessary connexion of any kind with
hypnosis. Thus, by neglecting the psychological effects of
suggestion, Charcot's demonstrations lost the reality of
scientific experiment for the appearance. An acrimonious
controversy ensued, but there is no doubt whatever that
Charcot was completely mistaken in his beliefs.

He was also wrong in another theory he put forward to
the effect that hypnosis could be induced only in abnormal
and neurotic subjects, particularly in hysterics, and that
hysteria was causally related to hypnosis. Bernheim retorted
that in his work, and in that of Liebeault, many hundreds
and thousands of entirely normal individuals had been
successfully hypnotized, and that consequently there was no
special relation of any kind between hypnosis and mental
abnormality. Again the so-called 'Nancy' school was right
and Charcot was wrong ; all modern workers are agreed that
the hypnotic trance can be produced at least as well in
mentally normal as in mentally abnormal individuals.

When we turn to modern theories of hypnosis, it cannot
be said that we leave the realm of absurdity behind. Some,
at least, of the more recent theories of hypnosis are equally
implausible as Mesmer's original notions, or Charcot's views.
A brief mention of some of the better-known ones may serve
to show the reader how very little agreement there is between
different authorities.

One of the older and more respectable theories sees in it a
modified form of sleep. The very term 'hypnosis' shows that
originally the sleep-like characteristic of the hypnotic trance

Hypnosis and Suggestibility 65

suggested an identification of the two states, and Pavlov
comes foremost in claiming that sleep and hypnosis are simi-
lar, involving a spread of cerebral inhibition in both cases.
This theory is almost certainly false. The physiological re-
action of the organism under hypnosis is quite different from
that which is observed in sleep. Thus, certain reflexes are
abolished in sleep, but not under hypnosis. Electro-
encephalogram recordings, or 'brain waves', show different
characteristics in the two states. The evidence is very strong
in opposing an identification of these two states.

A more acceptable hypothesis would regard hypnosis as a
conditioned response. Such a view might in due course be
elaborated into a proper theory, but at the moment it fails
completely to account for many of the phenomena associated
with hypnosis. How, one might ask, could conditioning ac-
count for spontaneous post-hypnotic amnesias? While con-
ditioning cannot be completely rejected as a likely part of a
true theory of hypnosis, certainly by itself it is not sufficient.

Much the same might be said of dissociation as an ex-
planation of hypnotic phenomena. It is well known that
parts of the cortex and the central nervous system can be
dissociated from the remainder, and many hypnotic pheno-
mena seem to be of this character. However, it will be diffi-
cult to account for hypnosis in terms of dissociation because
very little is, in fact, known about association, so that we
would merely be explaining one unknown by another.

A similar objection might be presented against another
view which looks upon hypnosis as an exaggerated form of
suggestibility. While, undoubtedly, there is a considerably
increased degree of suggestibility in hypnosis, it is idle to
seek for an explanatory principle in the laws of suggesti-
bility, because very little is known about suggestibility itself.
Again we would merely be attempting to explain one un-
known by another.

Among the more esoteric theories is a Freudian one, ac-
cording to which susceptibility to hypnosis depends on the
extent of ' transference ' formed between the subject and the
hypnotist. This 'transference' is a special relationship which

66 Part One • Borderlands of Knowledge

revives attitudes originally present in the parent-child rela-
tionship. Added to this are various erotic components which
are supposed to be present in hypnosis, which is considered
to be a manifestation of the Oedipus complex, of masochistic
tendencies, and so on.

Weirdest of all is a theory which states essentially ' that
the phenomena of hypnosis result from the subject's motive
to behave like a hypnotized person, as defined by the hypno-
tist, and as understood by the subject'. This is perhaps the
most question-begging of all, because it leaves unanswered
the two crucial questions as to why the subject should want
to behave in this fashion, and how he manages to do this. It
is all very well to say a person wants to behave like a hypno-
tized subject, but how does that help him to produce an
analgesia to an operation?

Most promising perhaps is a theory of ideo-motor action.
There is ample experimental evidence to show that ideas of
certain movements are closely related to the execution of
these movements. If electrodes and an amplifier are con-
nected to the muscles of the arm, and the subject is told to
lie quite still on a couch, but to imagine that he is lifting
that arm, then a barrage of nervous impulses is recorded as
passing through the nerves and into the muscles which
would have been used had the movement, in fact, been exe-
cuted. Thus, nerve transmission and mental images or ideas
are closely related, and, indeed, it appears that the one is
never found without the other. Without going into the ques-
tion of which causes which, the mutual interdependence of
mental and physical phenomena does not seem to be in
doubt. Under these conditions, the possibility of achieving
changes in a person's behaviour through verbal means, as in
hypnosis, appears possible. At best, this is only a partial
theory and it stands very much in need of considerable
amplification. If it could be combined with some such form
of inhibition theory as will be discussed in a later chapter,
we might here have the beginning of a true theory of hypno-
sis. At the moment such a theory cannot be said to exist, and
all that we can do is to note the experimental facts, which

Hypnosis and Suggestibility 67

are reasonably well established, and hope that a greater in-
terest in these important discoveries will eventually lead to
greater knowledge.

It is interesting to speculate why the development of a
scientific study of hypnotic phenomena has been so slow.
The main point undoubtedly has been the following. The
quotation is from Clark L. Hull, whose book on Hypnosis and
Suggestibility appeared in 1933, and which may be said to
mark the end of the pre-scientific and the beginning of a
truly scientific investigation of the subject. This is what he
has to say :

All sciences alike have descended from magic and superstition,
but none has been so slow as hypnosis in shaking off the evil asso-
ciations of its origin. None has been so slow in taking on a truly
experimental and genuinely scientific character The tardy de-
velopment of the science of hypnotism, moreover, is especially
striking when it is recalled that practically from the beginning
hypnosis has been definitely an experimental phenomenon. Not
only this, but experimentation has been continuous and wide-
spread during a period in which other fields of science have made
the greatest advances ever known. The paradox in this case, as in
all others, disappears with full knowledge of the attendant circum-
stances . . . the dominant motive throughout the entire history of
hypnotism has been clinical, that of curing human ills. A worse
method for the establishment of scientific principles among highly
elusive phenomena could hardly have been devised.... The
physician's task is to effect a cure in the quickest manner possible,
using more or less simultaneously any and all means at his dis-
posal. General laws which call for the varying of a single factor at
a time do not readily emerge from such situations.

The essential point emphasized again and again by Hull
is that what is lacking in most of the early work on hypnosis
is a notion of a controlled experiment. This notion is of al-
most universal application in science, and the method under-
lying it has been formally stated by John Stuart Mill in the
following words:

If an instance in which the phenomenon under investigation
occurs, and an instance in which it does not occur, have every

68 Part One • Borderlands of Knowledge

circumstance in common save one, that one occurring only in the
former ; the circumstance in which alone the two instances differ is
the effect, or the cause, or an indispensable part of the cause, of the
phenomenon.

As an example of the need for controlled experiments, let
us take the problem of the effectiveness of psychotherapy.
What most investigators have done in order to find out
whether psychoanalysis and other types of psychotherapy
are effective in producing a remission of neurotic symptoms
has been to take a group of seriously ill people, submit them
to the particular type of therapy in question, and state at the
end of a period of treatment, extending over several years,
how many of the patients have been cured, how many have
improved, how many remain in pretty much the original
state, and how many have actually deteriorated. This has
been the universal practice, but quite obviously it is not in
any sense conclusive. Suppose we find that 70 per cent of our
patients are cured after four years of treatment. This im-
provement in their condition might be due to the treatment,
but it might also be due to any number of other causes. We
can only be sure that it is the treatment which has been
effective by having a control group, i.e. another group of
patients, similarly afflicted, who do not obtain psycho-
therapy of the type administered to the experimental group.
If this control group fails to improve to the same extent as
the experimental group, then, indeed, we have reason to be-
lieve that psychotherapy has been effective.

In Uses and Abuses of Psychology I pointed out two facts.
The first was that no experiment along these lines had ever
been carried out, and that, of the fifty or so papers published
on the effectiveness of psychotherapy, not one had made use
of a control group. I also pointed out that it was possible
from certain reports in the literature to demonstrate that,
without any form of psychotherapy, considerable improve-
ment occurred in the condition of neurotic patients with the
simple passage of time. Indeed, this improvement can be put
in terms of a formula. If we let X stand for the per cent of im-
provement achieved, and jVfor the number of weeks elaps-

Hypnosis and Suggestibility 69

ing without treatment of a psychotherapeutic kind, then

jr=ioo(i-io-°-00435^).

Using these figures it was found that the improvement
claimed by psychotherapists after treatment was no more
and no less extensive than that found without any treatment
at all.

Since that time a proper experiment has been recorded
from California in which matched groups of neurotic sub-
jects were respectively treated by psychotherapy and not
treated at all. The outcome of the experiment was very
much in line with my previous conclusion. The treated
group improved to a considerable extent, but the untreated
group improved equally. Without the existence of such a
control group the erroneous impression would have been
given that psychotherapy was responsible for the improve-
ment. It cannot be emphasized too strongly that erroneous
conclusions of this kind have arisen so frequently in work on
hypnosis, on psychotherapy, and so forth, because of the
clinical inclination of the experimenters concerned. The de-
sire to help the unfortunate patient suffering from neurotic
and other symptoms has been stronger than the desire for
scientific knowledge, which alone can enable us to give
effective help. There are, admittedly, ethical problems in-
volved in withholding treatment from people in trouble, but
it should not be forgotten that there are also ethical prob-
lems involved in administering treatments the efficacy of
which is unknown, unproven, and suspect. Is the psycho-
therapist justified in demanding from his patient the con-
siderable expenditure of time, energy, and money, which is
involved in psychoanalysis when he has no evidence of any
kind to show that this treatment will be more effective than
doing nothing at all? Whatever the correct answer may be
to this problem, the reader may like to ponder the many
parallels and similarities between the development of Mes-
merism and that of psychoanalysis. In both cases there is
a strong personality as the founder of the cult; there is a
large congregation of pupils, fanatically devoted to the

yo Part One • Borderlands of Knowledge

furtherance of the master's teaching; there are the splits and
the formation of different schools ; there is a formulation of
unusual, unorthodox, and unlikely theories on the basis of
highly questionable evidence. In both cases there are re-
ports of cures achieved, and in both cases there is an absence
of the controlled experiment which alone could verify the
claims made. Only the future can tell whether the Freudian
libido will join animal magnetism on the heap of discarded
hypotheses for which science has no use.

2

LIE DETECTORS AND TRUTH DRUGS

If it be true that there are more things in heaven and earth
than are dreamed of in our philosophy, it is surely
equally true that things are dreamed of in our philosophy
which do not appear in heaven or on earth. Among these
figments of the imagination appear such varied objects as
the philosopher's stone, which was supposed to transmute
base metals into gold, the Oedipus complex, which was sup-
posed to transmute a normal person into a gibbering
neurotic, the houris, whose beauty and voluptuous sensual-
ity were supposed to console the Mohammedan warrior who
lost his life fighting for the faith of the Prophet, and the
Jungian archetypes, which are supposed to haunt our
modern minds with mystical reminders of the inherited wis-
dom, or otherwise, of our race.

A little nearer to a scientifically ascertainable fact than
these lie certain rather odd phenomena, which have re-
ceived a good deal of study in recent years under the general
title of psychosomatic disorders. What is meant by this
somewhat ill-conceived phrase is simply that certain dis-
orders of the body, or soma, may be caused by psychological
events, such as strong emotions, and that the cure for the
somatic disorder may thus be achieved by first accomplish-
ing a psychological purge. This insistence on the close rela-
tionship between body and mind, and the interplay be-
tween the two, is regarded as a very modern trend, and
Freud and the psychoanalysts generally are often credited
with the discovery of what many people have considered a
vital new truth in medicine.

This is a very unhistorical way of looking at the facts. The
general theory of psychosomatic interaction is at least as old
as human thinking about mind and matter, and there is
little in these modern theories that cannot be found in
the Greek philosophers or even earlier. Nor is the specific

7i

72 Part One • Borderlands of Knowledge

application of the principles involved to medical diagnosis
and treatment anything novel. To quote just one example,
we may consider a story told in the fifteenth century in a
well-known Persian work, the Akhlaq-i-Jalali. The author
tells how the great physician Rhazes was summoned to
Transoxiana to attend the Amir Mansur, who was suffering
from a rheumatic affection of the joints which none of his
medical attendants could cure. When he came to the river
Oxus, Rhazes refused to cross the river in the boat provided
because it was too small and fragile. The King's messengers
bound him hand and foot, threw him into the boat, and
carried him across by force. Rhazes explained the motive of
his resistance to them. He knew, he said, that every year
many thousands of people crossed the Oxus safely, but if he
had chanced to be drowned, people would have said what a
fool he was to expose himself to this risk of his own free will,
but had he perished while being carried across by force,
people would have pitied, not blamed him.

Having reached Bukhara, Rhazes tried various methods
of treatment without success. Finally, he said to the Amir,
'Tomorrow I shall try a new treatment, but it will cost you
the best horse and best mule in your stables'. The animals
having been placed at his disposal, Rhazes took the Amir to
a hot bath outside the city, tied up the horse and the mule,
saddled and bridled, outside, and entered the hot room of the
bath alone with his patient. He then took out a knife and be-
gan reviling the Amir, reminding him of the indignity he
had undergone in being carried across the Oxus by force,
and threatened now to take his life in revenge. 'The Amir
was furious and, partly from anger, partly from fear, sprang
to his feet.' Rhazes at once ran outside to where his servant
was awaiting him with the horse and the mule, rode off at
full speed, and did not stop in his flight until he had crossed
the Oxus and had reached Merv, whence he wrote to the
Amir as follows: 'May the life of the king be prolonged in
health and authority! Agreeably to my undertaking, I
treated you to the best of my ability. There was, however, a
deficiency in the natural caloric, and this treatment would

Lie Detectors and Truth Drugs 73

have been unduly protracted, so I abandoned it in favour of
psychotherapy, and, when the peccant humours had under-
gone sufficient coction in the bath, I deliberately provoked
you in order to increase the natural caloric, which thus
gained sufficient strength to dissolve the already softened
humours. But henceforth it is inexpedient that we should
meet.'

This example of 'psychotherapy' actually was designated
thus in the original tome, and turned out to be no less re-
warding on the material plane than modern psychotherapy
has been found to be. The Amir, delighted to find himself
restored to health and freedom of movement, awarded
Rhazes a robe of honour, a cloak, a turban, arms, a male
and female slave of beauteous countenance, a horse, fully
caparisoned, and further assigned to him a yearly donation
of 2,000 gold dinars and 200 ass-loads of corn.

Rather similar in principle is a story told by the greatest
of Mohammedan physicians, Avicenna, who was born
around the year 980. It comes from his rare and unpublished
Book of the Origin and the Return. The patient here is a woman
in the King's household who, while bending down to lay the
table, is attacked by a sudden rheumatic swelling of the
joints and is unable to assume an erect posture. The King's
physician, commanded to cure her, but having no medica-
ments at hand, has recourse to psychotherapy (again so called
by the author). He calls to his aid the emotion of shame and
begins to remove her clothes, starting with her veil and going
on to her skirt. Thus, in the author's words, 'A flush of heat
was produced within her which dissolved the rheumatic
humour', and she stood upright completely cured.

Many other stories in a similar vein could be quoted from
historical writings of many countries. They all suggest that a
knowledge of certain principles governing the relation be-
tween body and mind have been quite well understood from
time immemorial. One of these principles forms the basis of
our modern techniques of lie detection, and again we must
note that what appears to be a very modern development
can, in fact, be traced back hundreds and even thousands of

74 Port One • Borderlands of Knowledge

years to people whom we would regard as untutored
savages. One such application, vouched for by modern
anthropologists, goes back beyond the reach of human
memory. An example may illustrate the method used: A
chief of the tribe has been slain and five men to whom he
had done injuries in the past are suspected of having taken
revenge on him. How can the guilty one be found? The
tribe is drawn up in the form of a huge semi-circle by the
bank of the river; the five accused stand facing the tribe and
with their backs to the river. The witch doctor, gruesomely
attired and painted, is jumping about to the sound of
rhythmic drum-beats. The tension is growing steadily as the
moment of truth approaches. Finally, the dance comes to an
end and the witch doctor ceremoniously heaps rice from a
bowl on to five plates made of palm leaves. He then
harangues the tribe at great length about the iniquity of
the murder of a chieftain and about the magic which will
discover the murderer. Those who are innocent, he explains,
will eat the rice without difficulty; the guilty one, whose
crime is so repellent to nature that the very animals and
plants will have nothing to do with him, will find it impos-
sible to swallow even a mouthful and will thus stand re-
vealed in all his iniquity. Being a good practical psycho-
logist, as most witch doctors are (they would not survive
long if they were not!), he rams this suggestion home time
and time again, reciting previous trials where this infallible
method had been found to detect the culprit and produce a
confession. He then dramatically presents the five accused
with their platefuls of rice. Lo and behold, four of them pro-
ceed to eat it, if not with outward signs of enjoyment, at
least without any apparent trouble. The fifth, however,
ashen-faced, and hardly able to stand upright on trembling
legs, is moving his jaws desperately in a vain attempt to
swallow at least some of the rice, but without success. The
picture of guilt could not be clearer, and as he is dragged
away at the orders of the witch doctor, he bawls out his con-
fession before being thrown to the crocodiles.

The psychological processes at work here are reasonably

Lie Detectors and Truth Drugs 75

obvious. We all know the dry mouth of fear; strong emotion
inhibits digestion and also salivary secretion, which is so in-
timately connected with digestion. Without saliva, however,
the mastication and swallowing of food become difficult, if
not impossible. We can thus easily reconstruct what is hap-
pening to the unfortunate victim in the trial scene described
above. Knowing his guilt, and superstitiously afraid of the
power of the witch doctor and his ability to ferret out the
truth, he implicitly believes all that the latter is saying about
the difficulties the guilty person will experience in eating the
rice. Thus, fear of the consequences of what appears an in-
evitable discovery dries out his mouth, and the conscious
awareness of this change that is taking place further in-
creases his fear and his certainty of being discovered. When
finally the rice is actually produced he is in no fit state to
eat, and again his subjective awareness of the difficulties he
has in swallowing increases his fear, and thereby enmeshes
him further in the vicious circle of his own emotional re-
actions. Barbaric as this story may sound, there is nothing,
in principle, in modern methods of lie detection that is not
clearly contained in this account. We will find certain
technical improvements in recording the effects of emotion,
but, on the other hand, our modern procedures produce
much less emotion than the witch doctor was capable of
generating and, on balance, it is not certain that the advan-
tage lies with our modern techniques.

Salivation, as an index of emotion, is not used at all by
modern lie-detector experts. Nowadays we concentrate more
on the circulation of the blood, respiration, and certain elec-
trical phenomena of the skin, which will be described in due
course. A rather less gruesome story than the one about the
witch doctor may illustrate that these methods also were
well known a thousand years ago. It also concerns Avicenna
and is quoted in his own masterpiece, the Qanun, in the sec-
tion devoted to love, which he classes under cerebral or
mental diseases, together with somnolence, insomnia,
amnesia, mania, hydrophobia, melancholia, and the like.
When Mahmud of Ghazna was trying to kidnap him (kings

76 Part One • Borderlands of Knowledge

and rulers in general went to extreme lengths at that time to
obtain the services of a good consultant !), Avicenna fled and
came incognito to the town of Hyrcania by the Caspian Sea.
A relative of the ruler of that province was sick with a disease
which had baffled all the local doctors. Avicenna was asked
to give his opinion. Having examined the patient carefully,
he requested the collaboration of someone who knew all the
districts and towns of the province, and repeated their names
while Avicenna kept his finger on the patient's pulse. At the
mention of a certain town he felt a flutter in the pulse.
'Now', said he, 'I need someone who knows all the houses,
streets, and quarters of this town.' Again, when a certain
street was mentioned there was a flutter in the pulse, and
once again when the names of the inhabitants of a certain
household were enumerated. Then Avicenna said, ' It is fin-
ished. This lad is in love with such-and-such a girl, who lives
in such-and-such a house, in such-and-such a street, in such-
and-such a quarter of such-and-such a town; and the girl's
face is the patient's cure.' So the marriage was solemnized at
a fortunate hour chosen by Avicenna, and thus the cure was
completed.

Here, then, we again have an example of an involuntary
emotional response giving away a secret which, for some
reason or other, the subject of the experiment wants to keep
to himself, and again this involuntary response is one of the
well-known concomitants of strong emotion. This close rela-
tionship between the emotions felt by human beings and the
physiological changes taking place in their bodies is the
basis of our lie-detection techniques, and it has been dealt
with both on the experimental and on the theoretical level
by many psychologists and physiologists. Probably the most
famous is the American psychologist, William James, brother
of the novelist Henry James, who, together with a Nor-
wegian physiologist called Lange, gave his name to a law
which forms the basis, in one form or another, of most
modern work on emotions. This James-Lange law inverts
what we might consider the normal sequence of events.
What, in everyday opinion, happens when we feel an emo-

Lie Detectors and Truth Drugs 77

tion? We are sad and, therefore, v/e cry; we are afraid and
our heart beats faster; we are enraged and adrenal glands
pump adrenalin into our blood-stream. In other words,
emotion, as felt consciously by ourselves, comes first, the
physiological concomitants come second. James and Lange
maintain that this is putting the cart before the horse. In
response to a certain situation, the adrenal glands pour
adrenalin into our blood-stream, and therefore we feel
anger; in a certain situation our heart beats faster, and this
causes us to feel the emotion of fear; a certain situation
causes us to cry, and our subjective feeling response to the
crying is to feel sad. In other words, the external stimulus
situation (S) produces certain physiological responses (fears,
adrenalin, increase in heart-beat) which we may designate
PR. These physiological responses in turn produce the felt
emotion (E). Thus, the James-Lange formula is S^PR-^E,
whereas normally we conceive of the sequence as being
rather like this : S^E-^PR.

There has been much discussion of this theory and much
experimentation. Unfortunately, little of what has been
done can be said to throw much light on this problem,
which, in a nutshell, is a problem of the relation between
body and mind generally. We can only decide which causes
which - the emotion the physiological response, or the
physiological response the emotion - by finding out which
comes first, but this, unfortunately, we cannot do. Emotions
are subjective, and for timing we would have to rely on the
introspective reports of our subjects, which could not be ex-
pected to be very accurate, particularly when it is a matter
of split seconds. On the other hand, the physiological re-
sponse is not a sudden, all-or-none, affair; it also requires
some time to build up. These difficulties suggest that while a
decision may not perhaps be impossible, it is certainly im-
possible at the present time. Consequently, we need not con-
cern ourselves too much with this question. Let us merely
note that emotional experiences, as felt by the individual,
and certain types of physiological disturbances, as detected
by specialized electronic recording equipment, are always

78 Part One • Borderlands of Knowledge

found to go together, and that consequently we may use
evidence for the occurrence of the one as evidence for the
occurrence of the other. This essentially is the principle on
which lie detection is based, and there is no doubt that it is a
perfectly sound and respectable scientific principle. We will
have occasion to consider the difficulties which may arise in
using this principle a little later.

First of all, let us inquire what precisely are the physio-
logical reactions which denote the presence of an emotion.
Largely, these reactions can be identified because they are
transmitted by a special part of the nervous system. Speak-
ing quite broadly and generally, we may say that human
beings (and the higher animals as well, of course) have two
nervous systems. One, the so-called central nervous system,
is responsible for the transmission of impulses to the skeletal
musculature, which is responsible for carrying out voluntary
movements. Kicking a ball, writing a sonnet, jumping in the
lake, or putting a cross against a candidate's name, these are
all voluntary activities carried out by our skeleton, the bones
of which are being moved by muscles receiving their orders
from the cortex through the central nervous system.

There is, however, another system more ancient and rela-
tively independent of the central nervous system. This has
been called the autonomic, or vegetative, nervous system,
and deals essentially with certain vital but unconscious
activities, which maintain our body in good trim. Thus, we
breathe, our heart beats, digestion takes place, hormones are
pouring into our blood-stream, the amount of blood passing
through different parts of the body is finely regulated in re-
sponse to temperature, our pupil dilatates and contracts in
response to differences in brightness - all without any kind
of conscious adjustment. It is these autonomic, or vegetative,
responses which are so closely related to emotion. Some of
the major autonomic changes accompanying emotion are
familiar to everyone, and do not require instrumentation for
their detection. These changes include flushing, or pallor,
of the face, excessive sweating, increase in heart rate, the
drying out of the mouth, many vague visceral sensations,

Lie Detectors and Truth Drugs 79

and so forth. Under laboratory conditions many other more
subtle physiological changes can be observed, such as a rise
in blood pressure; increased oxygen consumption; dilatation
of the bronchioles of the lungs ; increase in the number of
red cells and platelets in the circulating blood ; liberation of
glucose into the blood-stream; secretion of adrenalin; reduc-
tion in the electrical responses of the skin; inhibition of peri-
stalsis in the gastro-intestinal tract; increase in blood-sugar;
and many hundreds of others which could be mentioned.

Most of these changes appear to have an adaptive pur-
pose; the major emotions, like fear and anger, are usually
the precursors of violent activity, which might be fight or
flight. For both a powerful supply of blood is needed. Conse-
quently, the heart beats faster to supply the anticipated
needs ; blood is withdrawn from the stomach, thus stopping
digestion ; some of the energy stored up by the body is re-
leased into the blood, which makes the organism more cap-
able of exertion. Cannon, in his book on The Wisdom of the
Body, has traced in great detail the wonderful adaptability
shown by our bodies in responding to urgent situations of
this kind; an adaptability which owes nothing to conscious
thought and everything to an inherited response pattern de-
veloped in millions of years of evolution.

Unfortunately, many of the emotional responses so pro-
duced are maladaptive in modern society. In civilized
society, neither fight nor flight are responses having survival
value. This generalization does not apply in exceptional cir-
cumstances, and in war-time particularly the responses
which were useful to our forefathers when they lived in the
jungle may become useful again. But even that becomes less
and less true as wars become depersonalized, and as hand-
to-hand combat is replaced by machine warfare. There is no
evidence to suggest that the bomb-aimer who releases the
unspeakable horror of the atom or hydrogen bomb by a
push of his finger on a button is enabled to do so with greater
precision because of strong emotion and vehement auto-
nomic activity. Quite on the contrary, it appears that fine
mental and muscular co-ordination, such as is required in

80 Part One • Borderlands of Knowledge

skilled tasks of this kind, is actually impeded by strong emo-
tion. Most people will have introspective evidence of this
from their own experiences. Clever argumentation becomes
disrupted by strong emotion; we cannot think as cogently
and as logically when we are annoyed as we do when we
consider the issues in an unemotional way. In playing a
game, strong emotion may actually make us more powerful
and less liable to feel injuries or succumb to exhaustion, but
it will also make us less skilful and less intellectual in our
approach; we may hit the ball harder, but accuracy will
suffer.

On the whole, then, strong emotions may prepare us for
the primitive type of battle in which brute strength and en-
durance and speed of flight determine survival, but they
will not only fail to be of advantage to us in most circum-
stances in the type of world we have created in the last few
hundred years - they will frequently be an actual handicap.
A struggle in the Board Room for control of a large enter-
prise, or a fight between rival politicians in an election; a
battle with the income-tax inspector for a change in one's
code number, or a negotiation with one's employers for an
increase in salary - all these are modern forms of combat
which have replaced the more physical types prevalent dur-
ing the course of human evolution. In all of them emotion is
a handicap rather than a help, and we have the unfortunate
position of having to reverse the evolutionary process and
try to respond to a challenge with an intellectual rather than
with an emotional response. It might not be too fanciful to
consider that our success in thus reversing the course of
evolution may determine our survival on this planet.

If this reversal were ever accomplished, human beings
would certainly make better liars. From many points of
view, of course, that would not altogether be an advantage,
and one of the socially useful functions of an emotion is cer-
tainly that of making lie detection possible. In a sense, of
course, the term 'lie detection' in this connexion is a mis-
nomer. What we, in fact, detect is the presence of some kind
of emotional response; we interpret this as evidence of lying

Lie Detectors and Truth Drugs 81

because we arrange the whole interrogation situation in such
a way that other sources of emotion are excluded as far as
possible. In this way, the fear of being found out, associated
with the telling of a lie, provides an emotional response
peculiar to the guilty person and is absent when the truth
is being told. Consequently, 'emotion detection5 becomes
'lie detection' in certain specially arranged circumstances,
but the jump from the one to the other still remains a jump,
and great care has to be taken, as we shall see presently, to
make certain that this jump is, in truth, justified and that
wrong deductions are not made from the truth as revealed by
the so-called lie detector. An example may make clearer
some of the difficulties encountered. Supposing that in the
case of the witch doctor's attempt to discover the guilty per-
son all five had been innocent, but one had known that the
witch doctor held a grudge against him and was going to
throw the blame for murder on him. Under those circum-
stances he might, quite reasonably, be expected to show the
same kind of fear reaction as the guilty person, and, conse-
quently, his failure to swallow the rice might be wrongly
interpreted as an indication of his guilt. There are several
possible ways of avoiding such pitfalls, and these will be dis-
cussed in the remainder of this chapter.

We must now turn to the instrument used in modern
work on lie detection. Popularly, the lie detector is often
thought of as an instrument which rings a bell or flashes a
light whenever a lie is told by the person who is undergoing
investigation. Unfortunately, nothing as certain and dram-
atic as this is available. Lie detection depends on deduction,
on circumstantial evidence, and on the fitting together of a
large number of disparate facts. It obtains its most depend-
able information from a continuous and simultaneous re-
cording of changes in blood pressure, pulse, and respiration
occurring during the interrogation of the suspect. These
recordings are made on an instrument, or polygraph, which
provides a progressive record of these various physiological
responses. The polygraph itself essentially consists of a long
roll of paper which is pulled across a surface at a uniform

82 Part One • Borderlands of Knowledge

speed, and on which a number of recording pens leave a
permanent record of their movements. The movement of the
recording pens in turn is governed by various instruments
attached to the person of the subject who is being investi-
gated. The instruments are connected to the recording pens
either electronically or mechanically.

Typical of the kind of instrument used is the so-called
pneumograph, which is used for recording breathing. The
pneumograph tube, which is tied round the subject's chest,
consists of a closely wound spring covered with a thin rubber
tubing; one end of the tube is sealed, while the other end is
connected to the polygraph itself by means of a small rubber
tube. During the test the circumference of the subject's chest
increases as he breathes in and decreases as he breathes out.
With each inspiration, therefore, the pneumograph tube
stretches, and with each exhalation it contracts. This move-
ment of the tube produces pressure changes inside it, and
these are transmitted to the polygraph and recorded there.

In an actual investigation our first task will be to obtain a
quiet, private examination room, to attach the various parts
of the instrument to the subject, and to inform him of the
general nature of the test. He is shown the instrument and
told that it is capable of determining whether or not a given
person is telling the truth. It is explained to him that the in-
strument records certain bodily changes which are indica-
tive of lying, but that it will, in no circumstances, cause
any kind of physical pain except for a slight and temporary
discomfort caused by the blood-pressure cuff. To reduce
nervousness and tension, he is told that if he is telling the
truth he has nothing to worry about because the instrument
will indicate that he is telling the truth. Further, he is told
that he will not be questioned about any of his personal
affairs, or about anything except the offence which is being
investigated. Everything is now ready for the test to begin.

The first part of the test serves very much the same pur-
pose as the long harangue by the witch doctor, describing
previous successes achieved by the use of his particular tech-
nique. The modern equivalent to the witch doctor is handi-

Lie Detectors and Truth Drugs 83

capped by the fact that his listeners would be very much
more sceptical and less liable to unreasoning fears and be-
liefs in his omnipotence than would primitive savages. Con-
sequently, instead of telling his subject what a wonderful
instrument this is, the modern operator would give an actual
demonstration. This is usually done in the following way.
The operator picks up seven or eight playing-cards and asks
the subject to pick out one, look at it, and return it to the
pack. He then tells the subject that he will show him each of
the seven or eight cards in turn and that he wants him each
time to answer 'No' to the question, 'Is this the card you
picked out?' He emphasizes particularly that even when the
right card comes along the subject should tell a lie and say
'No'. He then goes through the procedure of showing the
subject one card or another, eliciting a 'No' response to
each, all the time carefully watching the movements of the
recording pens on the polygraph. In ninety-live cases out of
a hundred a revealing over-reaction of the subject's auto-
nomic system is apparent in the recording when he answers
'No' to the card which he had seen before. To make quite
sure, the operator goes through the whole procedure again,
and then tells the subject which was the card he had looked
at. He also - and this is considerably more impressive to sub-
jects used to seeing this kind of thing done by sleight of hand
in the Music Hall - points out the actual changes taking
place in the polygraph recording when the lie is told. Thus,
the subject can see with his own eyes that reactions over
which he has no control can give him away completely, and
that the lie detector can be a very deadly instrument indeed
for anyone who is trying to give a false impression on any
matter of fact.

The subject, suitably impressed by now, is then asked a
series of questions relevant to the crime under investigation.
Depending on external circumstances, one of two rather
different techniques may be used. The first of these is the so-
called relevant-irrelevant question technique. In this, the
questions which are irrelevant to the point at issue, like ' Is
your name John Smith?', or 'Were you born in Liverpool?',

84 Part One • Borderlands of Knowledge

are asked in alternation with questions which are relevant
to the crime, such as 'Did you steal the diamond ring?', or,
'Did you shoot Lord Edgware?' This technique is rather
similar to that employed with the playing-cards, where the
card which the subject had looked at played the part of the
relevant questions, and the other cards would play the part
of the irrelevant questions. What is indicative of lying is a
difference in physiological reaction as we pass from the
irrelevant to the relevant questions.

What are the main changes which are indicative of de-
ception? The most reliable index is a simultaneous occur-
rence of a suppression in respiration and an increase in
blood pressure immediately after the subject's answer. How-
ever, even if only one of these two reactions occurs it is a
fairly safe guide to the examiner. Occasionally, deception
may be indicated by a decrease in blood pressure, occurring
several seconds after the subject has made his untruthful re-
ply to the question. Heavier breathing fifteen to twenty
seconds after a reply to a relevant question has been given is
also frequently indicative of lying; it is the physiological
concomitant of a feeling of relief that this dangerous point
has been passed with apparent safety, and it may also occur
at the end of the questioning period when the subject is told
that he will not be asked any more questions. A last decep-
tion criterion is a slowing up of a subject's pulse-beat imme-
diately after his reply to a question.

Certain rules must be followed in interpreting these re-
actions, however, the main rule being that in order to be
considered as evidence of lying, a physiological response to a
relevant question must be quite different to the physiological response
to an irrelevant question. Many people, although innocent of
the crime under investigation, have a bad conscience in
general, which makes them nervous and jittery altogether,
or are emotionally over-reactive. Such people show con-
siderable physiological reactivity, even after irrelevant ques-
tions, and might be considered to be lying if this fact were
not taken into account. The answer pattern on the irrele-
vant questions provides us for each subject with a base line

Lie Detectors and Truth Drugs 85

indicative of his response pattern when he is speaking the
truth ; lying can be diagnosed only when marked differences
from this normal response pattern are apparent in connex-
ion with relevant questions.

Another safeguard which should always be employed is
that no single response is taken as evidence of lying. The
same question, or a similar one, should always be asked
several times, and only if there is considerable consistency in
the subject's response pattern indicating his guilt should this
be regarded as suggestive. This is a very important safeguard
indeed. Accidental factors, such as a sudden cramp, a
sneeze, or a sudden loud noise, may produce effects indis-
tinguishable from the emotional concomitants of lying, and
these irrelevant factors can only be excluded by repeating
the whole procedure several times.

In certain circumstances a rather different form of
questioning can be used, which has great advantages. This
has been called the 'peak of tension', or 'guilty knowledge'
technique, and is dependent for its usefulness on the fact that
a guilty person may possess knowledge which no innocent
person would have. Any question regarding this knowledge,
or any reference to it, would produce emotional reactions in
the guilty person which would not be present in an innocent
one. An example from my own experience may make this
point clearer. At a certain hospital, bed-sheets were col-
lected once a week and put into large laundry bins on each
floor of the hospital. For several weeks running the sheets in
one of these bins were found mysteriously mutilated, and as
the patients had no access to the bins, suspicion centred on
about a dozen nurses working on that floor. No mention had
been made to them of these acts of vandalism and none of
them except the guilty one could therefore have known of
what had happened. This set the stage for the inquiry. The
nurses were told that they would be expected to take part in
a psychological experiment in which certain words would be
read out to them, and they had to say the first thing that
came into their minds. Among the hundred words used in
the experiment were a few which had some relevance to the

86 Part One • Borderlands of Knowledge

'crime'. Among these were bed-sheet, linen, cut, and bin.
After the records had been taken, these ' guilty ' words were
compared with the 'innocent' ones for each of the nurses.
Only one of them showed a very marked increase in auto-
nomic activity for the guilty words, and in her case this re-
action appeared to every one of the guilty words. On being
confronted with the record, she confessed and disclosed her
motives, which were essentially those of revenge against the
matron, who she thought had treated her badly.

In this particular case the type of physiological reaction
used was not one of those mentioned so far, but the so-called
psychogalvanic reflex. This is a rather mysterious pheno-
menon involving the measurement of the resistance of the
skin to the passage of an electric current. Electrodes are fixed
to the palm and the back of the hand, and a current so
slight as to be quite imperceptible is then set up. The resist-
ance offered by the hand to the passage of this current is
measured, and it has been universally found that any sud-
den shock or emotion produces a characteristic fall in this
resistance. This fall occurs after a latency period of about a
second or so after the stimulus has been applied and appears
to be proportional, roughly, to the amount of disturbance
caused. There is no universally accepted explanation of this
phenomenon, although there is some reason to believe that
it is connected with the activity of the sweat-glands of the
hand. Emotion is known to produce activity in these sweat-
glands, and sweat, being slightly salty, is a good conductor of
electricity. However, this is hardly the whole explanation,
and other possibilities have still to be considered. Fortun-
ately, the usefulness of this method of registering emotion
does not depend entirely on a knowledge of the mechanisms
which produce it and, in certain circumstances, the psycho-
galvanic reflex is a better indicator of deception than any
of those mentioned previously.

Curiously enough, the P.G.R. is hardly ever used in
actual police work. Apparently it provides too responsive a
measure of even slight emotional changes, so that in the
highly charged atmosphere of the police laboratory, re-

Lie Detectors and Truth Drugs 87

sponses to relevant and irrelevant stimuli may both show so
much evidence of emotion that differentiation becomes im-
possible. The P.G.R. comes into its own more in relation to
the parlour trick kind of investigation, as when the operator
wants to find out which of several cards the subject has
looked at, and under those conditions it is almost infallible.
It was used in the investigation described above because, to
the nurses, the whole experiment appeared as a kind of par-
lour trick and was not in their minds associated with the
question of crime, guilt, and lying at all. This may explain
the good results obtained here. For the interest of the reader
who may be unfamiliar with tracings of this kind, a record
is shown of the actual reactions given by the guilty subject
and by one innocent subject to the guilty words, and some
representative neutral words. In looking at this record, it
should be remembered that a dip in the curve is the equi-
valent to a decrease in the resistance of the skin to the pas-
sage of the superimposed electric current. There is little
need to comment on this record; it is literally true to say
that 'it speaks for itself (Figure 2).

This, then, is a rough picture of the kind of procedure em-
ployed in the lie-detector test and the kind of criteria used for
establishing the presence or absence of emotion accompany-
ing a reply. How valid is the lie-detector test, and how much
reliance can be placed on it? These are difficult questions,
and before answering them we must consider certain factors
which affect the interpretation of scores. One of the main
difficulties in tests of this kind is the nervousness shown by
many a truthful or innocent person under the conditions of
police interrogation. Such nervousness is usually manifested
in the record by the general erratic character of the tracings,
and by the failure of the larger reactions to be related speci-
fically to the pertinent questions; thus, the physiological dis-
turbances associated with nervousness appear on the lie-
detector record without any consistent relationship to any
particular question or questions. They are no greater when
pertinent questions are asked than when neutral questions
are asked for control purposes. Sometimes, changes or

88 Part One • Borderlands of Knowledge

disturbances of a similar size to those appearing in the record
are even found during rest periods when no questions are
being asked at all. The best way of counteracting the pre-
sence of nervousness in a subject is by reassurance and by
repetition. Repeating the whole procedure five, or even ten,
times acts very much like a sedative because the subject,

w

Tabic Sheet Face Shirt Linen, Shoe

See

<f /f .f ^ ^ f.

Table 5heet Fare shirt Linen shoe

w

See Cut Dream Light Bin Flower

Figure 2: (a) P.G.R. reactions of guilty person; 'guilty'
words underlined. Arrow indicates when this stimulus
word was pronounced, (b) P.G.R. reactions of innocent
person; 'guilty' words underlined

who may originally have been afraid of all sorts of things
(many people when they are being strapped into the ma-
chine express fear of receiving electric shocks, for instance),
now gets used to the procedure and realizes that there is
little in it of which to be afraid. Also, frequent repetition, as
mentioned before, enables the examiner to look for consist-
ency of exaggerated responses to the same question, or ques-

Lie Detectors and Truth Drugs 89

tions which are regarded as relevant, and this consistency is
far more important than individual isolated reactions can
ever be.

Nervousness is relatively easily diagnosed (it does not
usually require an inspection of the lie-detector record to
tell whether or not the person is extremely nervous!), and it
does not present an insuperable difficulty. Other factors are
more difficult to deal with. Mental deficiency, for instance,
makes a subject almost useless for a lie-detector test. A very
stupid person, who may not be able to distinguish between
truth and falsehood, or understand the social obligation of
telling the truth, or who has no fear of detection, will not
give useful results which can be interpreted in any way.
Much the same is true of young children, who also are not
fit subjects for lie-detector tests. Obviously innocent persons
frequently give records so abnormal and erratic that no
interpretation is possible. Again, these various conditions do
not present insuperable difficulties. Mental deficiency and
mental abnormality sufficient to invalidate the lie-detector
test are easily recognized, and other methods of arriving at
the truth substituted for the lie detector.

Much greater difficulties are caused by certain people
who are relatively unresponsive to the conditions of testing.
A person who refuses to believe in the efficacy of the test,
and who therefore is not worried at all about giving himself
away, will tend to show less emotion when telling a lie and
therefore produce less easily interpretable records on the
polygraph. Some people show a quite extreme absence of
emotion, and such people, while actually guilty, might
easily be pronounced innocent in terms of their reaction in
the lie-detection test. Again, some people are able to control
their mental associations and processes sufficiently to avoid
give-away responses. An excellent example of this technique
occurred in connexion with the trial of Jerry Thompson,
who was executed for the rape-murder of Mildred Hall-
mark. His lie-detector records were rather indefinite and
could not be interpreted. However, quite independently of
the lie-detector test, he confessed, and several days before

go Part One • Borderlands of Knowledge

his execution was interviewed by the lie-detector expert,
who was interested to learn how Thompson had been able
to beat the test. Thompson stated that whenever he was
asked during the test whether he ' raped and killed Mildred'
he would concentrate upon, and enact in his own mind,
various abnormal sexual experiences he had had with an-
other girl of the same name. He stated that by doing this
each time the question was asked he was able to dismiss
for the time being the rape and murder of Mildred Hallmark
from his mind. Fortunately, only few people seem to be able
to control their minds to the extent required in order to
'beat the machine', but it should be remembered that some
people do possess this ability, and that their number seems
to be particularly large among hardened criminals.

Among persons knowing the physiological basis of the lie
detector, other tricks may sometimes be employed to make
detection impossible. It will be remembered that lie detec-
tion depends very much on a comparison between items
which are relevant and items which are irrelevant. If the
subject is capable of causing physiological reactions to irre-
levant items, then the comparison will become useless and the
record will not be capable of revealing damaging informa-
tion. Causing physiological responses to innocent items may
be done in various ways. One subject, a physiologist, who
had pushed his wife down into the Grand Canyon, produced
autonomic reactions when being asked irrelevant questions
by stepping on his left foot with the heel of his shoe, the re-
sulting pain producing a reaction sufficiently like those of
lying to make the record unusable. Other people have been
known to indulge in sexual fantasies when answering inno-
cent questions, thus producing physiological responses to
the fantasies which the investigator would attribute to the
next question. It is not categorically possible to say that such
tricks may not work occasionally, but in most cases the ex-
perienced examiner will notice what the accused person is
doing and will draw his own conclusions.

We are now in a position to assess the practical utility of
lie-detector tests. By and large, it is difficult to disagree with

Lie Detectors and Truth Drugs 91

Inbau, a very experienced expert in this field, when he
writes that

Lie-detector tests, when conducted by competent and experi-
enced operators, are of considerable practical utility. In the first
place, with the aid of a lie-detector, it is possible to detect decep-
tion with much greater accuracy than is otherwise attainable.
Secondly, the instrument, the tests, and the accompanying pro-
cedures have a decided psychological effect in inducing admissions
and confessions from guilty individuals.

Unfortunately, exact figures as to the accuracy of lie-
detector results are almost impossible to obtain. In actual
police work, confirmatory or contradictory evidence cannot
always be obtained, so that we may never know in certain
cases what the truth of the matter really was. For this rea-
son, many investigators have given figures purporting to
demonstrate the accuracy of lie-detector tests which are
based on experiments conducted in the laboratory. In these
experiments no actual crime is committed, but the task of
the investigator is to find out, as in the example already
quoted, which of several cards an individual has looked at,
or, alternatively, which of several individuals has read a cer-
tain letter. In such laboratory tests the actual truth is known
and the success of the lie-detector tests can be checked
against it. Reports of accuracy varying from 80 to 100 per
cent have been published, and it would probably be a fair
estimate to say that, when properly conducted, such tests
are accurate in about 95 per cent of the cases.

Unfortunately, we cannot extrapolate from experimental
investigations of this type to the kind of work going on in the
police laboratory. It has already been pointed out that
whereas in the psychological laboratory the psychogalvanic
reflex type of response is particularly effective, it is by no
means so useful in the interrogation by the police of actual
suspects. Why this should be so is not too clear at the present
moment, but the fact must be faced that the accuracy of
police work cannot be gauged adequately by reference to
work in the psychological laboratory7.

92 Part One • Borderlands of Knowledge

The most convincing estimate of the accuracy of lie-detec-
tion techniques comes from Inbau, and is based on some
twelve years experience in the Scientific Crime Detection
Laboratory in Chicago. According to his estimate, out of
ioo cases the examiner can make a definite and accurate
deception diagnosis in about 70. In twenty cases the records
will be too indefinite in their indications to permit a com-
petent and cautious examiner to make a definite diagnosis.
Doubt in these cases may be due either to the inconsistency
and the rapid nature of responses in the record, or else to the
general unresponsiveness of the subject. As regards the re-
maining ten cases, here even the most experienced examiner
is likely to make a definitely erroneous diagnosis. The chief
source of error in these cases lies in the failure of the examiner
to detect deception in a guilty individual rather than in his
misinterpretation of the record of the innocent subject.

This is a conservative estimate, and it should be said that
some experts have published claims for their methods of 97,
98, or 99 per cent accuracy, or even in one case for 100 per
cent efficiency. These are undoubtedly exaggerations, and
the disappointment inevitably following expectations based
on such claims has probably been responsible for some of the
distrust of lie-detector techniques, which can be observed
among some members of the police force.

At first sight the degree of accuracy achieved by the lie
detector in competent hands may not seem impressive. If
we leave out the 20 per cent of cases where the experts can-
not come to a decision, we find that in seven out of eight
cases the expert is right, leaving a margin of error of some-
thing like 12 per cent. To anyone desirous of 100 per cent
accuracy - and in science this degree of accuracy must be
the goal, however much we may in practice fall short of it -
a margin of error of 1 2 per cent may appear discouragingly
high. However, while it limits the usefulness of the lie de-
tector, it certainly cannot be said to render it useless.

In the first place, one of the most valuable by-products of
the lie-detection technique is that it helps to obtain a con-
fession in cases where other methods have failed, and where

Lie Detectors and Truth Drugs 93

a considerable amount of work would be required if the con-
fession had not been forthcoming. To be confronted with the
objective evidence of his lying on the tracing paper of the
polygraph is a very disconcerting experience for most
criminals, and experience has shown that confessions are
produced in a very large number of cases. Indeed, there
have been innumerable instances of confessions made as a
result simply of the proposal to have a suspect submit to a
lie-detector test. On many occasions suspects have confessed
their guilt while waiting to be tested. Others, again, have
confessed immediately after the operator had set up the test
and adjusted the instruments.

Confessions so obtained still have to be checked, of course,
against objective evidence; a confession alone, whether ob-
tained with or without the use of the lie detector, would not
be regarded by most courts as sufficient. However, included
in the confession are usually details which make it possible
to obtain an objective verification of the claims of the guilty
subject, and it should certainly be stressed that so far no case
has come to light where false confessions have been pro-
duced because of the psychological effect of the instrument
or the techniques used in administering it. In this there is a
very marked difference between lie-detection techniques,
and the use of the 'third degree', which has caused many
subjects to confess to crimes they have not committed in
order to escape from an intolerable situation.

This ability of the lie detector to induce confessions in
guilty persons is, to many people, its main claim to practical
usefulness. However, in spite of its liability to error in a small
portion of the cases, we cannot ignore the contribution
which, in fact, is made by the lie detector, even in cases
where no confession is obtained. To have a guide to the
truth, even if it be only a 90 per cent reliable guide, is cer-
tainly better than to have no guide at all. In many cases, the
lie detector test record enables the police to abandon a line
of inquiry because of the obvious innocence of the subject
under suspicion. In other cases, the suspicion of his guilt,
strongly supported by the lie detector test record, may

94 Part One • Borderlands of Knowledge

enable the police to concentrate efforts on the most promising
suspects. In many cases, details of the examination may be
helpful in suggesting certain clues to the police, such as the
names of likely confederates, or the place where money or
murder weapons were hidden, and so forth. In all this, of
course, it is important to realize that conviction is not based
directly on the record of the lie detector. This record is only
one item which is taken into account in arriving at a con-
clusion; one item of evidence among other items, none of
which can be said to be perfectly reliable, but all of which
together point to the guilt or innocence of a given suspect.

People who criticize the use of lie-detection techniques on
the basis of this lack of perfect reliability often overlook the
fact that all other techniques at present in use for determin-
ing the truth or falsity of a suspect's story are at least equally
fallible, and in many instances more so. The question that
should be asked is not whether a new departure, such as the
lie-detection technique, is completely accurate, but rather
whether it is more accurate than the techniques which it
would replace, and whether, in fact, its employment would,
or would not, improve the degree of success obtainable at
the present moment. On these grounds there is little doubt
that the lie-detection technique emerges with flying colours.

To these considerations, arguing in favour of the use of
lie-detection techniques, there should also be added an-
other. The only possible way in which imperfect techniques
can be improved is by experimentation and use. To say that
lie-detection techniques should not be used because they are
not perfectly reliable at the same time means that the only
path leading towards an improvement of the lie-detection
technique is blocked. It is possible, up to a point, to conduct
experiments in the psychological laboratory, but for the
practical development of a technique such as that of lie de-
tection, actual use in practice is indispensable. Only by such
use can the most effective and efficient method of interroga-
tion be worked out; only in this way can competent per-
sonnel be trained ; and only thus can theoretical research be
directed along the most promising paths.

Lie Detectors and Truth Drugs 95

To say that lie-detection techniques should be used in
police work is rather a vague statement. There are many
ways in which a technique such as this can be used, and it is
important to be more precise in one's recommendations.
Before making these recommendations we may perhaps look
at the present legal studies of the lie detector. I have been
unable to find any legal decisions regarding its use in Great
Britain, and we must therefore perforce turn to the United
States, where several courts have considered this issue. The
first appellate court decision upon the admissibility of lie-
detector evidence was given in 1923 by a Federal court. The
accused, on trial for murder, offered as evidence the results
of a lie-detector test. The reasons why this offer was refused
were very clearly stated : 'Just when a scientific principle or
discovery crosses the line between the experimental and de-
monstrable stages is difficult to define. Somewhere in this
twilight zone the evidential force of the principle must be
recognized, and while courts will go a long way in admitting
expert testimony deduced from a well-recognized scientific
principle or discovery, the thing from which the deduction
is made must be sufficiently established to have gained
general acceptance in the particular field in which it be-
longs. We think the systolic blood pressure deception test
has not yet gained such standing and scientific recognition
among physiological and psychological authorities as would
justify the courts in admitting expert testimony deduced
from the discovery, development, and experiments thus far
made.' Several other cases have come up in more recent
years, all agreeing in refusing to admit the evidence of lie-
detector tests. Consequently, evidence of this kind has prac-
tically no judicial recognition at the present time in the
United States, and it is probably safe to say that the position
in Britain would not be different.

Are these decisions reasonable, or are they simply evi-
dence of an ' old-fogey ' mentality, so frequently found in
traditional pursuits like the law? Here again there is little
doubt that we must agree with Inbau, the expert whose
estimate about the reliability of lie-detection results we

96 Part One • Borderlands of Knowledge

quoted before. This is his conclusion from considering the
evidence :

It is generally recognized that our present legal methods and
procedures for ascertaining the truth and administering justice are
far from perfect and that therefore we should forever remain alert
and eager for corrective measures and improvements. At the same
time, however, we should not be too hasty in our acceptance of
proposed innovations. In the case of the lie-detector technique such
a cautious attitude will operate not only in the best interests of the
administration of justice but also to the ultimate advantage of the
technique itself. A premature acceptance of the test results as legal
evidence would undoubtedly occasion such a series of abuses and
miscarriages of justice as to stigmatize forever the technique in the
field of law as well as of science. It seems much wiser, therefore, to
await a further development and improvement of the technique,
and to let it first stand the acid test of a truly scientific inquiry into
its accuracy and reliability.

Inbau bases this conclusion on two main arguments. A
technique, in order to be admissible, must possess ' a reason-
able measure of precision in its indications'. The percentage
of errors admitted to be in the neighbourhood of 10 per cent
is no greater, and in many cases no less, than that possessed
by some forms of scientific testimony actually admitted by
the courts. However, it should be remembered that if lie-
detector test results were to be admitted as legal evidence,
they would be offered and treated as proof of some im-
portant phase of the case, or even of the validity of the en-
tire claim or contention of one of the parties. In other words,
a decision on the truth or falsity of a statement given in
answer to the question, 'Did you shoot X?' would com-
pletely decide the case. This is not so with most forms of
scientific testimony. A ballistics expert might be prepared to
give evidence, saying that a given bullet had been fired by a
given gun, but this single fact would not, in itself, decide the
issue. Quite generally, scientific testimony is not ordinarily
conclusive regarding the entire issue under litigation, and
consequently there is not so much likelihood of injustice re-
sulting from erroneous expert testimony. Lie-detector evi-

Lie Detectors and Truth Drugs 97

dence, however, usually relates to the crux of the whole mat-
ter, and consequently, if it were admitted, an error would be
extremely serious. The words, 'a reasonable measure of pre-
cision', therefore have to be interpreted in the context of the
importance of the decision; what might be a reasonable
measure of precision in a small matter might be quite un-
reasonably lacking in precision, when it is a matter of life
and death. As it is in the latter kind of case that appeal is
made to the lie-detector test, its precision and accuracy re-
quire to be greater rather than less than those of other forms
of scientific evidence.

When it comes to the second prerequisite as to the ad-
missibility as legal evidence, namely, the acceptance of a
technique in the fields of science to which it belongs, we find
that while the principles underlying lie-detection techniques
are recognized by physiologists and psychologists, this par-
ticular form of application of the principle is not so widely
accepted. In part, this is due to ignorance; few psychologists
and physiologists have taken an interest in this matter, and
consequently, quite rightly, are not prepared to commit
themselves on a subject-matter in which they are not them-
selves experts. In addition, the exaggerated and sensational
claims made by a few lie-detector operators have made
scientifically minded people very chary indeed of accepting
any claims at all in this field.

In addition, the lie-detection techniques at present in use
have not been adequately standardized with respect to in-
strumentation, the manner in which a test should be con-
ducted, the interpretation of the recordings, or the training
of competent examiners. In such circumstances, incom-
petent or dishonest persons might represent themselves as
'lie-detector experts' and be allowed to give inaccurate or
perjured testimony for the party by whom they are em-
ployed.

There is one exception to this rule of inadmissibility of lie-
detector test results in court. If the lawyers representing the
prosecution and the defence agree on admitting lie-detector
test results as evidence, and also agree on a particular expert

98 Part One • Borderlands of Knowledge

to carry out the investigation, such evidence has been recog-
nized, at least in the United States, by a number of trial
courts. This exception is justified for two reasons. In the first
place, it may be assumed that when both the opposing liti-
gants and their lawyers are willing to resort to the use of lie-
detector tests, the case must be a doubtful one, i.e. one where
the evidence on either side is unconvincing and incapable of
circumstantial corroboration. (Sexual crimes often fail into
this category, as in most cases none but the two persons in-
volved are present, and physical evidence as to precisely
what happened is rarely forthcoming.) In those circum-
stances, where any decision reached on the basis of the avail-
able evidence would be little more than guess-work, there is
little doubt that the utilization of lie-detector test results
would substantially increase the accuracy of the final de-
cision over what it would have been had it been based
merely on the guess or hunch of judge and jury deciding in
the absence of lie-detector evidence. There is also the point
that, where both the parties agree on the selection of the
test expert, he is likely to be both honest and competent and
not in any way influenced by an attachment to one or the
other party.

It may appear to the reader that in what we have said so
far there is a glaring contradiction. On the one hand, it has
been maintained that the present practice of not making the
lie-detector test results legally admissible evidence is a rea-
sonable one, and that at present there are no grounds for
altering this position; on the other hand, it has been sug-
gested that official use should be made by the police of the
lie detector. How can these apparently two contradictory
recommendations be reconciled? The answer, in effect, has
already been given when it was pointed out that one of the
main advantages of the use of the lie detector was the pro-
curement by its use of confessions in cases where the more
usual types of police interrogation had failed. Confessions,
admissions, and other evidence obtained in consequence of
the use of a lie detector are admissible in court and have
legal status, at least in the United States; the legal admissi-

Lie Detectors and Truth Drugs 99

bility of such evidence in Great Britain has not so far been
tested, but it seems likely that the position here would not be
very different from that obtaining in the United States. A
legal decision on this point was made by the Supreme Court
of Pennsylvania in 1939 in a case in which the defendant
confessed to having committed a murder. This confession
followed the administration of a lie-detector test, and the
defendant's counsel objected to the admissibility of a confes-
sion. The objections were based on the fact that a lie detec-
tor had been used, and the fact that the investigating officers
had told the suspect that * You can lie to us, but you cannot
lie to this machine'. The Pennsylvania Supreme Court up-
held the trial court's ruling that the confession was admis-
sible. They stated that since no promises, force, or threats
had been employed in obtaining the confession, the mere
use of the instrument did not render it inadmissible. Nor
would it regard the officers' comments about the impossi-
bility of lying to the machine as invalidating the confession.
Even though these words might be considered a trick, that
fact alone would not invalidate the confession, because of the
general legal rule in the United States which makes admis-
sible confessions procured 'by a trick or artifice not calcu-
lated to produce an untruth'.

Another argument sometimes used by lawyers appearing
for the defence is that the use of the lie detector constitutes a
' third-degree' practice. This argument has not recom-
mended itself to most competent observers or legal authori-
ties. Some slight discomfort of a purely temporary nature is
produced by the blood-pressure cuff, but this is much too
slight to be considered painful in any sense. Furthermore,
the test procedure is not of a kind which would encourage or
compel a person to make a confession merely in order to
escape from an intolerable situation. It is interesting to note
in this connexion that there are several cases in which a sus-
pect, having withstood the rigour of third degree interview-
ing without confessing, has finally admitted his guilt after a
brief session with the lie detector. The use of the third de-
gree, apart from being uncivilized and non-humanitarian,

ioo Part One • Borderlands of Knowledge

is also inefficient, and a substitution of the lie-detector test
in its place in American police interrogations would con-
siderably improve the efficiency and accuracy of crime
detection there.

This whole question of using tricks in order to obtain a
confession or admission from a criminal raises many thorny
legal problems. It is not at all certain that to say 'You can
lie to us, but you cannot lie to this machine ' is a mere trick,
because, in fact, the statement is substantially correct. Of
course, in a sense this might be regarded as a quibble be-
cause the statement becomes true only by being made in the
first place, and by being believed by the suspect. However
that may be, many people, particularly those of a humani-
tarian cast of mind, and with a high degree of religious or
ethical responsibility, have argued against the use of lie-
detection techniques because of a feeling that it weights the
scale somewhat unfairly against the accused and makes use,
if not of trickery exactly, then of something very closely ap-
proaching it. With this general view, which, in fact, was re-
sponsible for the elaboration of the rules of evidence which
the police are obliged to follow in Great Britain, one cannot
but have much sympathy. To make sure that the underdog
is fully protected against society, and that injustice is not in-
flicted on him because he cannot stand alone against the
forces representing the law, this is one of the most important
features of democratic life, and one which few people would
be willing to give up in the interests of greater efficiency.

However, if the underdog is to be protected from society,
particularly in its monolithic modern form, society also has
to be protected from the vicious criminal, and one's sym-
pathy should go, not only to the wrongdoer, but also to his
actual or prospective victim. Take the case of Jerry Thomp-
son, whose trial for rape-murder caused quite a sensation a
few years ago. Thompson had a long history of assault and
rape. In this particular case Thompson had choked and
slugged his victim in a fit of passion, as he had done many
times before in connexion with other girls. In this particular
instance, however, his victim failed to recover consciousness

Lie Detectors and Truth Drugs 101

soon enough and, assuming her to be dead, he threw her
body from his automobile. Apparently she was still alive
when thrown from the moving car and only died from the
impact her body made on the road.

A lie-detector interrogation technique was used on
Thompson, who later confessed. Few people, surely, would
consider that morality would have been satisfied more by
allowing him to go free and pursue his career of assault and
rapefor several more years rather than makehimsubmit him-
self to a lie-detector test and be made to suffer the conse-
quences of his actions. This may be an extreme example, but
it may counteract our sentimental tendencies to see vicious
and brutal killers and murderers as essentially kind-hearted
underdogs who never had a fair chance.

Essentially, surely, the position is this. The lie-detector
test in itself is legally neutral. Its efficacy at the present time
is reasonably well known, and its value in producing true
confessions is without doubt. It does not in any way partake
of 'third-degree' methods; it does not cause physical pain to
the accused; and it does not, when it fails, fail by implicating
the innocent person. Its failure rather is a failure to detect
the person who has lied. Lastly, in many cases it serves as a
shield to the innocent person, who, without fault of his own,
is enmeshed in a mass of circumstantial evidence, but the
truth of whose protestations of innocence is verified by the
lie detector. It would be foolish to accept testimony so
secured without reservation, to admit it to the status of legal
evidence, and to employ the technique in every single case.
It would be equally foolish to refuse to recognize its potential
usefulness, to refuse to employ it in carefully selected cases
where it would show to maximum advantage, and to refuse
to use it with strict safeguards in the furtherance of the ends
of justice.

The whole position was put in a nutshell by none other
than Daniel Defoe, who in 1 730 published a pamphlet called
An Effectual Scheme for the Immediate Preventing of Street Rob-
beries and Suppressing all Other Disorders of the Night. As the fol-
lowing quotation will show, Defoe not only discovered the

102 Part One • Borderlands of Knowledge

basis of the modern lie-detector technique, but also dealt
with the ethical objections to its application very much in
the way that modern writers do.

Guilt carries Fear always about with it ; there is a Tremor in the
Blood of a Thief, that, if attended to, would effectually discover
him; and if charged as a suspicious Fellow, on that Suspicion only
I would always feel his Pulse, and I would recommend it to Prac-
tice. The innocent Man which knows himself clear and has no Sur-
prise upon him; when they cry, Stop Thief, he does not start; or
strive to get out of the Way; much less does he tremble and shake,
change Countenance or look pale, and less still does he run for it
and endeavour to escape.

It is true some are so harden'd in Crime that they will boldly
hold their Faces to it, carry it off with an Air of Contempt, and out-
face even a Pursuer ; but take hold of his Wrist and feel his Pulse,
there you shall find his Guilt; a fluttering Heart, an unequal Pulse,
a sudden Palpitation shall evidently confess he is the Man, in spite
of a bold Countenance or a false Tongue : This they cannot con-
ceal; 'tis in vain to counterfeit there; a conscious Heart will dis-
cover itself by a faltering Pulse; the greatest Stock of Brass in the
Face cannot hide it, or the most Firm Resolution of a harden'd
Offender conceal and cover it: The Experiment perhaps has not
been try'd, and some may think it is not a fair Way, even with a
Thief, because 'tis making the Man an Evidence against himself:
As for that, I shall not enter into the Enquiry farther than this; if it
is agreeable to Justice to apprehend a Man upon Suspicion, if the
Particulars are probable and well grounded; it cannot than [sic]
be unlawful by any Stratagem that is not injurious in itself, to seek
out collateral Grounds of Suspicion, and see how one thing concurs
with another.

It may be true, that this Discovery by the Pulsation of the Blood,
cannot be brought to a Certainty, and therefore it is not to be
brought into Evidence ; but I insist, if it be duly and skilfully ob-
served, it may be brought to be allow'd for a just Addition to other
Circumstances, especially if concurring with other just Grounds of
Suspicion.

In concentrating in this chapter so much on the lie detec-
for, and in devoting only a few words to the discussion of
'truth drugs', we may be said to give an accurate presenta-
tion of the importance of these two types of development and

Lie Detectors and Truth Drugs 103

their prospective usefulness. The lie detector is based on a
well-established scientific theory; it is of acknowledged use-
fulness, and its value is vouched for by impeccable scientific
research. Truth drugs, on the other hand, are in quite a
different category. Their value is exceedingly problematical ;
their use is based on no well-thought-out scientific theory.
Only in one respect do they resemble the lie-detector test,
and that is in having a very long history, stretching back
into primitive times. Even among ancient Romans the ten-
dency of alcoholic beverages to cause people to reveal what
they would much rather have kept secret was well known,
and the proverb in vino Veritas has its analogue in every
civilized tongue.

Modern truth drugs are similar to alcohol in this, that
they depress the activity of the higher centres of the brain,
thus temporarily releasing the lower centres from surveil-
lance. It is in these unguarded moments, when the ever-
wakeful censor is, as it were, drugged, that certain state-
ments may slip out which otherwise would have been
severely blue-pencilled. The balance, however, is difficult to
achieve. A little too much and the lower centres also become
paralysed, and the subject goes to sleep; too little and the
higher levels retain their censorship functions and little
worth-while is revealed. Even when the proper balance is
struck, however, and conditions are optimal, it is doubtful if
the truth is really revealed. By this, I mean that an indivi-
dual conscious of a certain fact, and having strong motives
for not admitting such knowledge, could not, in effect, be
made to reveal it by any of these so-called truth drugs at pre-
sent available. This was shown dramatically in an experi-
ment carried out on a variety of normal and neurotic sub-
jects. They were told a story and also told that they must not
reveal any details of this story to a person who would come
and question them about it. Injections of pentothal (one of
the truth drugs) were then administered, and an attempt
made to elicit details of the story from the subjects. The
effort ended in failure. Normal people did not, in fact,
divulge any information at all; the neurotic ones did spill

104 Bart 0ne ' Borderlands of Knowledge

details of the story, but they introduced so much imaginary
material that it was quite impossible to reconstruct the truth
from their outpourings. On the whole, then, there is no good
evidence to suggest that a conscious determination not to
divulge a secret can be overcome by the use of drugs at
present known to medical science. The possibility that there
may be secret drugs available to certain Eastern govern-
ments and unknown here cannot, of course, be ruled out,
but in the absence of any evidence it would require a con-
siderable amount of credulity to take such hypotheses very
seriously.

There is, however, one use of the so-called truth drugs
which does lend some point to the use of that term. Neurotic
patients are often unconscious of certain events in their past
lives, and such events are frequently of great emotional im-
portance and closely related to the particular disability
from which the patient suffers. Thus, a soldier suffering from
a hysterical paralysis may have acquired this symptom as a
direct outcome of being buried as an effect of an exploding
shell; he may have completely forgotten this episode and
have no memory of it at all. When questioned under one of
the truth drugs, he frequently will not only remember the
episode, but may actually relive it all, cowering in the corner
of theroom, shouting with fear, and finally subsiding into un-
consciousness. The therapeutic use of such a drug to induce
recollections cannot easily be assessed, but the fact that fac-
tually true statements are made under such conditions of
events which the subject does not recollect in the normal
state cannot be doubted.

An example may make clear the difference between the
recovery of repressed material of this kind of which the sub-
ject is himself unconscious, and the discovery of certain facts
which are well known to the subject, but which he is not
willing to disclose. During the war a middle-aged soldier
was picked up by the military police, wandering about Lon-
don. He claimed an almost complete amnesia, could not re-
call his name, his civilian address, or, in fact, anything about
his circumstances at all. He carried no marks of identifica-

Lie Detectors and Truth Drugs 105

tion, and it proved impossible to discover anything about
his true identity. After an injection of pentothal he still
failed completely to remember anything at all, and this
roused considerable suspicion, as a true neurotic amnesia
could have been expected to show some signs, at least, of
breaking down under the drug. However, about the same
time he fell in love with a young female patient at the hos-
pital, and they were about to get married when, one day,
going down the local High Street, he encountered a woman
accompanied by seven children who claimed to be his wife
and would not let him go back to the hospital. Her story
turned out to be true, and after some questioning the man
finally admitted to having made up the whole story of his
amnesia in order to get away from his wife.

From the point of view of police interrogation and getting
the truth from someone determined not to reveal it, truth
drugs are of little value. It is possible, but not very likely,
that new discoveries in the near future will improve the
situation. Until more is known about the physiological basis
on which these drugs work, it is doubtful if very much useful
experimentation can be done in this field. Possibilities exist, of
course, of combining truth drugs and lie-detector techniques,
but unfortunately the truth drugs themselves have a direct
influence on the autonomic nervous system, thus counter-
acting the smooth running of the lie-detection technique.
For the moment, at least, we may dismiss the idea of a ' truth
drug', properly so called, from our minds and concentrate
for all practical purposes on the development of the lie
detector.

3

TELEPATHY AND CLAIRVOYANCE

According to T. H. Huxley it is a customary fate of new
truths to begin as heresies and to end as superstitions. While
many people would not be willing to regard psychical
phenomena in any sense as 'truths', they certainly have be-
gun as superstitions, and belief in them at the moment is re-
garded as a heresy in scientific circles. Thus, in this as in so
many other ways, telepathy and clairvoyance seem to be
intent on standing the ordinary, accepted, and understand-
able course of nature on its head.

The types of phenomena with which psychical research
deals have been known for many thousands of years. They
include such things as premonitions, fire-walking, water-
divining, haunted houses, poltergeists, survival after death,
and telepathic messages. In the words of an official descrip-
tion of the aims of the Society for Psychical Research, it is
the object of the research worker in this field 'To examine
without prejudice or prepossession and in a scientific spirit
those faculties of man, real or supposed, which appear to be
inexplicable on any generally recognized hypothesis.' This
definition is a little sweeping. We have no recognized hypo-
thesis to account for the apparent interaction of mind and
matter in a simple act of consciousness, nor is there any
official hypothesis to account for the phenomena of hypnosis
or of memory. Yet consciousness, hypnosis, and memory are
not generally included in psychical research.

We will not attempt here any general definition of psychi-
cal phenomena, but will merely indicate the nature of those
phenomena which will be dealt with in this chapter. These
are often subsumed under the general term 'extra-sensory
perception', or E.S.P. for short. By this is meant the ac-
quisition of knowledge through channels other than those of
the senses. Two main types of extra-sensory perception are
recognized, namely, clairvoyance and telepathy. Clairvoy-

106

Telepathy and Clairvoyance 107

ance is supposed to occur when a person experiences a cer-
tain mental pattern which corresponds wholly, or in part,
with the sensory aspects of a past, present, or future physical
object or event, in such a way that the observed correspond-
ence cannot be accounted for by sense perception or infer-
ence based on sense perception or by chance coincidence. A
further requirement, the necessity for which we shall see a
little later, is that this physical object or event is never at any
time perceived by anyone. (This added requirement may
seem to make it quite impossible to check up on the occur-
rence of clairvoyance, but it will be seen later that ingenious
methods of experimentation have succeeded in overcoming
the apparent difficulties involved in this condition.)

Telepathy is supposed to occur when a person experi-
ences a mental pattern which wholly, or in part, corre-
sponds to a past, present, or future mental pattern of an-
other person, whether living or dead, under conditions when
the correspondence cannot be accounted for by sense per-
ception or inference based on sense perception or by chance
coincidence. A further requirement is that the possibility of
clairvoyance should have been eliminated. In the great
majority of studies which have been reported, it is impos-
sible to decide whether the E.S.P. faculty supposedly de-
monstrated is telepathic or clairvoyant, because the facts, if
they be accepted as such for the moment, can be accounted
for equally easily in terms of either. Consequently, for the
most part, we shall be concerned with the generalized extra-
sensory perception faculty rather than with any demarca-
tion between clairvoyance and telepathy. There are, how-
ever, one or two specialized studies which have attempted
rigorous discrimination between these two faculties, and
these will be introduced in their appropriate place.

One further phenomenon will be dealt with briefly, and
that is a faculty or ability called psycho-kinesis, often short-
ened to PK effect. As telepathy and clairvoyance derive
from ancient claims of sorcerers, mediums, and medicine
men, so psycho-kinesis derives from poltergeists and the be-
liefs of gamblers that they can influence the fall of dice by

108 Part One • Borderlands of Knowledge

willing them to fall in a certain way. The alleged ability of
human beings to influence the behaviour of physical objects
without the transfer of any known form of energy is the
subject-matter of psycho-kinetic studies, and although only
little is known about it, it cannot be wholly omitted in a
chapter such as this.

The notion of investigating claims of this kind will be re-
pugnant to many people. The very possibility of extra-sen-
sory perception, or psycho-kinesis, appears contrary to
modern scientific logic, and many people have shown con-
siderable reluctance even to look at the evidence that has
been produced in favour of these alleged abilities. Many lay-
men have a kind of stereotyped view of the scientist as an in-
human, completely objective and rational sort of person,
who only takes into account facts and is not swayed by emo-
tions and feelings in his judgement. Unfortunately, there is
little truth in such a picture. Scientists, especially when
they leave the particular field in which they have special-
ized, are just as ordinary, pig-headed, and unreasonable as
anybody else, and their unusually high intelligence only
makes their prejudices all the more dangerous because it en-
ables them to cover these up with an unusually glib and
smooth flow of high-sounding talk. We shall encounter a
good deal of such talk in the course of our exploration.

How, it might be asked, did this type of research origin-
ate? The answer is that it originated in much the same way
as any other kind of scientific research. Certain phenomena
having been reported consistently for thousands of years,
certain obvious problems are being raised. At face value
they seem to contradict certain widely held beliefs; yet it
seems difficult to dismiss these phenomena as being simply
due to chance, to misconception, or to fraudulence. In other
words, we seem to have a problem here, and a particularly
important and interesting one, because it touches on the
root of our modern scientific outlook on life. Small wonder
that scientists with an open mind began to apply the
methods of science to this field in order to throw some light
on these interesting and unlikely events.

Telepathy and Clairvoyance 109

Let us take just a few well-known examples from ancient
and modern times to illustrate the kind of phenomena which
aroused this type of interest. St Augustine, who must be re-
garded as a reasonably trustworthy witness, relates that one
of his pupils asked Albicerius, a Carthaginian medium and
diviner, to say what he, the pupil, was thinking about.
Albicerius replied that the pupil was thinking of a line of
Virgil, and although he was a man of little education, actually
recited the passage. This correct perception on the part of
Albicerius of what was in the mind of another person would
suggest the reality of telepathy, although, of course, there may
be more acceptable ways of accounting for the coincidence.

Another example is the case of Sosipatra, a feminine don
and philosopher in ancient Greece, who interrupted a lec-
ture she was giving to describe in detail an accident that was
happening at that very moment to her kinsman Philometor,
who was riding in his carriage many miles away. Her de-
scription appears to have been correct in most of its details,
and again an explanation in terms of clairvoyance or tele-
pathy is the most obvious, although, of course, not a neces-
sary explanation of this event. Better known, perhaps, than
either of these stories is that of King Croesus, as told by
Herodotus. Croesus sought the help of a medium to dis-
cover the strength of his enemies. Being experimentally
minded, he first of all decided to find out which was the
most reliable of the many oracles. Accordingly, he sent his
representatives to the seven most famous ones, instructing
them to make their visits all on the same day and to put to
all the oracles the same question : ' What is the King of Lydia
doing at this moment?' When the specified day arrived, the
king went through the most improbable performance he
could think of. He cut up a tortoise and a lamb and boiled
them in a brass cauldron. Only the famous oracle at Delphi
was able to give the correct answer. When the King then
asked what would happen if he attacked the Persians, he
was told that a mighty empire would be destroyed. He ac-
cepted this as an encouragement, and, indeed, the predic-
tion was fulfilled in the event, but it was his own empire

no Part One • Borderlands of Knowledge

that was destroyed. We hardly need extra-sensory percep-
tion of the precognitive kind to account for the success of the
oracle at Delphi's prediction because, in a war, one or the
other side must, after all, win, but the divination of what the
king was doing is rather more difficult to explain.

It is, of course, possible in all these stories that rumour,
and the constant retelling, have added many details which
make the whole thing more miraculous than it might have
been to begin with, and it is difficult to rule out coincidence
and trickery completely. It would be more difficult, how-
ever, to use such interpretations of certain more modern
events. Here, for instance, is a story told by Dr J. F. Laub-
scher, a psychiatrist working in South Africa. He describes
how he made a friend of a native witch doctor who was be-
lieved to have E.S.P. abilities. On one occasion, without
mentioning his plans to anyone, Laubscher secretly hid a
cheap purse in the ground and then went at once by car to
the witch doctor's kraal some 60 miles away. In the course
of a seance dance, the witch doctor was able to describe cor-
rectly, in minute detail, the hidden article, down to the
colour of the wrappings and the nature of the locality where
it was buried.

Stories of this type are not, of course, acceptable as proof,
however miraculous they may seem. They pose a problem
of explanation, but no scientist would rate their value any
higher than that. There are too many uncontrolled features
about them, and these make it unsafe to take them very serious-
ly. However, for several years after the founding of the Society
for Psychical Research in London, an event which took
place in 1882, the main energies of the investigators were
concerned with the collection of stories of this type, as well
as the actual investigation of cases of psychical phenomena
reported to them. Interesting as much of this early work is,
it again fails to satisfy the demand for proof. Too much at-
tention is paid to human testimony, and relatively uncon-
trolled conditions are accepted too readily. Human testi-
mony, unfortunately, is so unreliable, and the human being
is an observer so much given to inaccuracies and downright

Telepathy and Clairvoyance 1 1 1

falsifications, that little credence attaches to anything re-
ported by human beings having nothing to aid them but
their senses. That this is not too harsh a verdict has been
shown in any number of experimental investigations. These
have mostly been concerned with the fallibility of testimony
given before courts of law, but their results are very relevant
here also. In the usual experiment, a group of students is
assembled, receives a lecture on the inaccuracy of testimony,
and then some kind of complex and exciting event is stage-
managed. An intruder may come in, brandish a revolver,
threaten to shoot the lecturer, and then be frog-marched out
by two of the students. Written accounts are then called for
of what has happened, and few people who have not par-
ticipated in experiments of this type would credit the variety
of inaccuracies appearing in these reports. The revolver be-
comes a knife, or a gun, or a stick; the intruder becomes a
woman, an old man with a beard, a Negro, or a lunatic. His
words are completely garbled and he may, in the stories,
either run away by himself or be marched out by dozens of
students.

A good example of this technique of investigating the re-
liability of reports is an experiment reported by S.J.Davey.
He was interested in the kind of phenomena reported during
seances and, using quite simple trickery, which he had
planned in advance, he reproduced some of the effects popu-
lar among the mediums of the day. His audiences were
asked to write down accounts of what they had witnessed,
and these observations were then compared with what
actually happened. Here is a report written by one witness
of such a seance. ' On entering the dining-room where the
seance was held', so the report runs, 'every article of furni-
ture was searched and Mr Davey turned out his pockets.
The door was locked and sealed, the gas turned out, and
they all sat round the table holding hands, including Mr
Davey. A musical box on the table played and floated about.
Knockings were heard and bright lights seen. The head of a
woman appeared, came close and dematerialized. A half-
figure of a man was seen a few seconds later. He bowed and

ii2 Part One • Borderlands of Knowledge

then disappeared through the ceiling with a scraping noise.'
Another witness also described the searching of the room,
the sealing of the door, and the disposition of the medium
and sitters round the table. She alleged that a female head
appeared in a strong light and afterwards a bearded man
reading a book, who disappeared through the ceiling. All
the while Mr Davey's hands were held tightly by the sitters
on either side, and when the gas was relit the door was still
locked and the seal unbroken.

A third witness's account was even more sensational. He
reported that ' nothing was prepared beforehand, the seance
was quite casual'. Having described the locking and sealing
of the door, he went on to say that he was touched by a cold,
clammy hand and heard various raps. After that he saw a
bluish- white light which hovered over the heads of the sitters
and gradually developed into an apparition that was
'frightful in its ugliness, but so distinct that everyone could

see it The features were distinct ... a kind of hood

covered the head, and the whole resembled the head of a
mummy'. After this an even more wonderful spirit appeared.
It began with a streak of light and developed by degrees
into a bearded man of Oriental appearance. His eyes were
stony and fixed, with a vacant, listless expression. At the end
of the seance the door was still locked and the seal was in-
tact.

So much for some of the reports. Now for the reality. The
seance was not a casual affair at all, but had been carefully
rehearsed beforehand. At the beginning, Mr Davey went
through the motion of apparently locking the door, but he
turned the key back again so that the door was actually left
unlocked. The 'props' for the materializations had been
stowed away in a cupboard underneath a bookshelf; this
was not looked into by the witnesses who searched the room
because, just as they were about to do so, Mr Davey diverted
their attention by emptying his pockets to show that he had
nothing hidden on his person. The phenomena were pro-
duced by a confederate who came in by the unlocked door
after the lights had been turned out, and while the musical

Telepathy and Clairvoyance 113

box was playing loudly to drown the noise of his entry. The
'apparition of frightful ugliness' was a mask draped in mus-
lin with a cardboard collar coated with luminous paint. The
second spirit was the confederate himself, standing on the
back of Mr Davey's chair, his face faintly illuminated by
phosphorescent light from the pages of a book he was hold-
ing. The rasping noise made when the spirits seemed to dis-
appear through the ceiling was caused accidentally, but
interpreted by the witnesses according to their conception
of what was happening. When the light was turned on the
gummed paper that had been used to seal the door had fallen
off, but Mr Davey quickly pressed it back into position and
then called the witnesses' attention to the fact that it was
'still intact'. Mr Davey's performances were so convincing
that some leading investigators, including the biologist A. R.
Wallace, f.r.s., refused to believe him when he said that he
had no mediumistic powers and it had all been done by
trickery. In effect the conjurer was challenged to prove that
he was not a medium !

More recently other investigations of a similar kind have
been reported, and they all agree that, from the point of
view of proof, human testimony in the absence of careful
controls is quite useless. This conclusion will hardly surprise
anyone who has ever seen a good conjurer at work on the
stage. Few of the happenings in seances are anything like as
miraculous as what happens on the brightly lit stage in front
of thousands of people. A few hours' instruction in element-
ary conjuring should enable any reasonably adept person to
produce most of the alleged psychical phenomena seen at
seances.

It should be stated quite emphatically that the witnesses
who were misled in these experiments were not particularly
foolish or particularly credulous people. It is very doubtful
if, under conditions of this kind, even the most eminent
scientist would have been any more likely to give a correct
account. It is generally recognized in science that instru-
ments are required to supplement available human observa-
tion, and that years of training in the use of such instruments

ii4 Part ®ne * Borderlands of Knowledge

are essential before reports can be accepted as representing
the truth of what actually happened during the experiment.
The notion that psychical phenomena could be investigated
without such safeguards is a slightly absurd one because, if
anything, the conditions surrounding these observations
(darkness, excitement, etc.) are such as to make the usual
safeguards more rather than less necessary and desirable.

Another difficulty which must be taken into account is the
fact that even when a highly intelligent and apparently
trustworthy witness is speaking about events which have
happened to himself, he may be giving an account which is
far from true. A good example of such fabrication is the case
of Sir Edmund Hornby. Sir Edmund Hornby was Chief
Judge of the Supreme Consular Court of China and Japan.
According to his story, which was published by the Society
of Psychical Research in 1884, he stated that a certain news-
paper editor and reporter used to call at his house in the
evening and collect his written judgements for the day so
that they could be printed in the next morning's paper. One
night Sir Edmund was roused by a tap on the door, and in
walked the reporter. Looking deadly pale, he approached
the foot of the bed, and in a polite but desperately insistent
manner pleaded with Sir Edmund to give him verbally a
summary of the judgements. Sir Edmund complied, and the
reporter thanked him and left the room. Sir Edmund looked
at the clock, which showed the hour to have been half past
one. Lady Hornby woke up and was told what had hap-
pened.

The next day, Sir Edmund heard that the reporter had
died in the night. He had been seen writing at a quarter to
one, and at half past one his wife found him dead. Beside his
body was his shorthand notebook with the heading for the
judgements the last item in it. He had apparently died of
heart disease, and it was absolutely impossible for him to
have left home during the night.

Here, then, is a highly impressive story, told by a witness
of unimpeachable integrity. Yet it emerged later that the
editor in question had died between eight and nine o'clock

Telepathy and Clairvoyance 115

in the morning on a day following one on which no judge-
ments had been made in court, and three months before Sir
Edmund had married. On hearing these facts, which proved
indisputable, Sir Edmund admitted that his memory must
have played him an extraordinary trick. It is unlikely that
he would have invented the story and allowed it to be pub-
lished, knowing all the while that it was false and apt to be
exposed. It appears more likely that he did have a vision of
the dead reporter, but that the various striking details were
added to the occurrence bit by bit, until finally his memory
of the event had become almost completely changed. Such
falsification is liable to occur in one's recollection of any
highly emotional scene, and again it should not be assumed
that Sir Edmund Hornby reacted in any way differently
from how other people might have reacted. Testimony of a
single person, however authoritative he may be, must never
be accepted as evidence for a psychic experience.

Must we, then, reject all these earlier accounts of
mediums, seances, and telepathic and clairvoyant perform-
ances? The answer is that the great majority are suggestive
but could never be used to convince the sceptic, and in a
new field such as this, the scientist must almost by definition
assume the role of sceptic. If any alternative explanation to
extra-sensory perception is possible, then that alternative ex-
planation, however unlikely it may be, must be accepted.
Thus, as long as there is any possibility of fraud, trickery, col-
lusion, coincidence, memory falsification, or anything else
of this sort, the investigator may regard the reported events
as interesting, and may use them to suggest hypotheses to
him, but he must not, under any circumstances, regard them
as evidence in favour of the occurrence of either telepathy or
clairvoyance.

There are, however, a few events which are rather diffi-
cult to explain away, even by the most sceptical. As an
example, let us take the case of Mrs Piper, a young married
woman of Boston, Massachusetts. Her abilities were inves-
tigated by Professor William James, one of the really great
names in psychology. Knowing the quality of most mediums,

1 1 6 Part One • Borderlands of Knowledge

James was inclined to scoff when two of his female relatives
were greatly impressed by Mrs Piper. He went to see for
himself and quickly changed his mind. She became his 'one
white crow', and throughout her career she was constantly
being investigated by highly critical and able experimenters.
James himself was joined in this work by Dr Richard Hodg-
son, Secretary to the American Society for Psychical Re-
search. Hodgson had quite a reputation for exposing fraudu-
lent mediums, and he went so far as to set detectives to
watch Mrs Piper and her family to find out if she made
secret inquiries about her sitters' private affairs. Neither the
detectives nor any of the investigators ever discovered Mrs
Piper in a dishonest act.

Mrs Piper's forte was to tell strangers things about their
private affairs which she could not normally have known. In
order to make ignorance even more certain, Mrs Piper was
invited to England so that she could meet foreign sitters,
about whose personal affairs she could not possibly have any
prior information. Here is an example of the type of work
she did. She was staying at the time with Sir Oliver Lodge,
Professor of Physics at Liverpool University, in his own
house. All the servants were new and ignorant of the family
connexions, and Lodge, very sceptical at this stage, took
such precautions as locking up the photograph albums and
family Bibles, and searching Mrs Piper's luggage. Strangers
were asked to call and were introduced to Mrs Piper under
assumed names. Mrs Piper herself was quite uninquisitive
and seemed somewhat self-absorbed.

Lodge tried an experiment with Mrs Piper. He wrote to
an uncle to ask for a relic of the uncle's twin brother who
had died about twenty years previously. He was sent an old
watch, which he gave to Mrs Piper while she was in a trance.
She said, almost at once, that the watch belonged to an
uncle, and after a lot of stumbling she produced the name of
'Jerry'. Lodge encouraged ' Uncle Jerry ' to recall boyhood
incidents that his surviving brother would remember.
Several such incidents were mentioned, including the
swimming of a creek, being nearly drowned, killing a cat in

Telepathy and Clairvoyance 117

Smith's field, and possessing a long peculiar skin, like a
snakeskin. The uncle with whom Lodge was in correspond-
ence did not recollect all these points, but on writing to yet
another brother, Frank, verification was received for every
one of the items mentioned by Mrs Piper.

Lodge sent an inquiry agent to Uncle Jerry's home village
to see how much information could be obtained by question-
ing the old men of the place about Smith's field and the rest.
The result was almost nil. It is particularly interesting in
this connexion to note that Lodge himself did not know any-
thing about Uncle Jerry's boyhood and therefore could not
say until after he had made inquiries whether or not the
medium's statements were correct. Here, it seems, we have
a case reliably reported by several people of integrity and in-
telligence where explanations not involving extra-sensory
perception are rather difficult to come by unless, indeed, we
explain it simply in terms of coincidence. Yet coincidence
would not often result in such startling revelations, and it
should be remembered that Mrs Piper did not produce this
as an isolated event, but kept on producing 'coincidences',
for some twenty-five years, day in and day out. There are
one or two other mediums whose performances rival those of
Mrs Piper, and while scientists will always tend to have a
rather odd feeling at the very mention of the term ' medium',
this does not justify us in rejecting the performances so care-
fully studied and investigated.

Yet however unlikely ' coincidence ' may sound as an ex-
planation of the events ascribed to Mrs Piper and other
mediums, we cannot set an actual probability level to the
likelihood of such events as the name of Lodge's uncle being
correctly produced by Mrs Piper. What appears unlikely to
one person may not appear so unlikely to another, and
science does not deal with subjective estimates of this kind.
We must have some firm statistical estimate of the proba-
bility of correct guesses, as well as the total number of
guesses produced, before we can make any rational and
scientifically acceptable statement about matters of this
kind. The odd and uncontrolled events of seances, of

1 1 8 Part One • Borderlands of Knowledge

prophetic dreams, of the utterances of mediums - all these
may be lively and interesting and intriguing, but they cannot
easily be subjected to the processes of counting, of measure-
ment, and of statistical probability theory.

The main contribution of psychology to the study of
psychical phenomena has been two-fold. In the first place,
the circumstances in which these phenomena are pro-
duced have been placed under the control of the experi-
menter, and in the second place the phenomena have been
made amenable to measurement and statistical calculation.
This two-fold process of tightening up conditions of the ex-
periment and of obtaining an exact statement of the proba-
bility of ' coincidence ' as an explanation has made it pos-
sible to achieve a degree of rigour which makes the accept-
ance or the rejection of the evidence no longer a matter of
individual belief. While this change in emphasis has been all
to the good and has, indeed, been absolutely essential for
making the alleged phenomena acceptable to scientists, it
has also had one rather unfortunate effect. Gone are the
miraculous stories and the interesting personalities, the
mediums, the conjurers, the oracles. Gone are the impas-
sioned arguments about trustworthiness and about the
effects of chance. Gone are the serious, long-bearded Vic-
torians, crowding into small rooms to watch ectoplasm
emerge from the mouths of mediums. Instead of all this
colour and glamour and amusement we now have clean-
shaven young scientists repeating hundreds of thousands of
times simple card -guessing games ; we have calculating ma-
chines churning out probabilities in terms of highly complex
formulas which only the initiated can comprehend. The
whole colourful story is boiling down to simple statements,
such as that Mr A, in five million calls, in which by chance
he should have been correct once in fifty-two cases in calling
the right card, was correct once in fifty cases, the probabili-
ties against this happening by chance being said to be
astronomical.

There is little point in regretting the passing of the Vic-
torian era of psychical research; nothing we can do will

Telepathy and Clairvoyance 119

bring it back. But the reader will soon notice that the im-
provement in methodology has to be bought at a price, and
this price, to many people, has been the unmitigated bore-
dom produced by having to read through countless reports,
all as similar as two peas, of people doing exactly the same
thing under exactly the same conditions, getting results
slightly better than chance would allow. However, fortun-
ately, even in this bleak field there are items of interest and
moments of excitement. Before turning to these, let us con-
sider quite briefly the statistical basis of modern research on
extra-sensory perception. Let the reader take a pack of fifty-
two playing cards, shuffle it thoroughly, cut it, and look at
the top card. As he is looking at the card, let him press an
electric key which gives a signal to a person in another room
to write down his guess at the suit of the card. The pack is
then shuffled again after the card has been replaced, and
card after card is drawn, looked at, and replaced in the same
manner. Let us make 1,200 guesses in this way. Now it will
be clear that the probability of a guess being right by chance
is one in four. Consequently, if nothing but chance were
operating in our experiment, then the probability of a cor-
rect guess should be one-quarter of 1,200, or 300 in all. Of
course, in any actual experiment it is unlikely that the num-
ber of guesses would be precisely 300. We would have some-
times a few more and sometimes a few less. If we had 100
experiments, each consisting of 1,200 trials, and recorded
the mean number of successes for each of these experiments,
we would find that the scores, i.e. the number of successes,
would vary around the 300 mark from about 270 to 330.
Very occasionally we would get more extreme scores, but
the very great majority, on the hypothesis that chance only
was operating, would be within that range.

Suppose, now, that we are testing a subject in this manner
whose score over the 1,200 trials is 400. Can we dismiss that
as a simple coincidence? The answer is not simply 'Yes' or
'No'. What we do is to calculate the probability of such an
extreme deviation occurring by chance. In our case this
probability would be less than a million to one and we

120 Part One • Borderlands of Knowledge

would certainly be justified in thinking that chance was a
pretty poor explanation. If the chances of a horse winning
the Derby were a million to one against, few people would
be willing to bet on that horse, and similarly, if the chances
against the 'coincidence' hypothesis are a million to one, as
in this case, few scientists would be willing to back it. Thus,
we would be able to conclude that something was probably
acting here other than chance to produce the observed
results.

If we were very strongly opposed to a belief in telepathy
or clairvoyance, we might say that the odds against it were
so strong that no single experiment could be considered con-
vincing. We would then simply have to duplicate the experi-
ment ; if we obtained another run of 400, the probabilities
against chance being active would be something like a mil-
lion times a million. We could go on piling probabilities on
probabilities, but for most people a reasonable level like a
million to one would probably be acceptable enough, pro-
vided the experiment was technically satisfactory in other
respects. Of course, the reader might insist on adopting
Pascal's logic to this problem. Pascal, it will be remem-
bered, was a very gifted mathematician, as well as a very
bigoted Catholic. As befits one of the founders of modern
probability theory, he used his theory to prove that we
should believe in God. His argument runs as follows: The
felicity of Paradise is infinite. Therefore, however low the
probability be that God exists, if you multiply this proba-
bility by infinity it becomes certainty. Consequently, we
ought to believe in God. Pascal's argument appealed neither
to the mathematicians nor to his co-religionists, and has re-
mained somewhat of a historical curiosity. Its inverse, which
states that psychical phenomena are infinitely unlikely and
that, therefore, no finite set of probabilities can prove them,
has received somewhat wider acclaim, although its logical
basis is equally fallacious.

Many people, of course, believe that you can prove any-
thing by statistics and are, consequently, neither impressed
nor surprised by these modern findings. This is rather too

Telepathy and Clairvoyance 1 2 1

drastic a view to take, because the greater part of modern
science is based on statistical methodology and the calculus
of probability ; to relinquish all of modern science in order
to get rid of the necessity of having to admit psychical
phenomena is like throwing out the baby with the bath-
water. It may be possible to prove anything by statistics,
but only to people who are foolish or ignorant; the method-
ology of modern statistics is so widely accepted by scientists
that few people who know anything about it would imagine
that they could get away with any kind of fallacious proof
simply because the proof rested on a statistical basis.

The essence of modern work in psychical research, then,
consists in arranging the experiment in such a way that the
probability of any given result occurring by chance can be
calculated, thus giving us an estimate of the likelihood of
'coincidence' having been responsible for our findings. Even
if ' coincidence ' is to be ruled out as a likely cause of our
findings, this does not, of course, by itself prove the existence
of extra-sensory perception. There may be weaknesses in the
experimental design itself which are responsible for the
extra-chance results we obtained. Before we can turn to
statistics as the arbiter, we must ensure that not even the
most persistent and sceptical critic can find anything to cavil
at in our experimental design. Only then does the statistical
argument become of crucial importance.

The first experimenter whose work should be mentioned
here was an American psychologist, Coover, who was
strongly opposed to the theory of extra-sensory perception.
He used two people for each of his experiments, the 'sender'
or 'agent', who looks at the cards one by one, and the 're-
ceiver' or 'recipient' who records his guesses. Coover used
altogether 105 guessers and ninety-seven senders. He sat
with the sender in one room while the subject being tested
sat in an adjoining room. The purpose of having two rooms,
of course, is to make it absolutely impossible for the receiver
to catch any glimpse of the cards, and to prevent the sender
giving him any signal, intentional or unintentional, which
might enable him to identify the card by normal means.

122 Part One • Borderlands of Knowledge

Altogether, 10,000 guesses were recorded, using packs of
playing cards from which the twelve picture cards had been
removed, so that each pack contained forty cards. Having
shuffled and cut the pack, Goover threw a die for ' odds ' or
c evens ' in order to decide whether or not the sender should
look at the face of the card drawn. Coover intended those
trials in which no one saw the card to act as an empirical
control on the trials in which the agent concentrated on the
cards, but nowadays we could consider them rather as clair-
voyance trials, as opposed to the telepathy trials.

Goover reported that neither the telepathy trials nor the
' control ' trials showed any significant deviations from mean
chance expectation, and he accordingly concluded that
there was no evidence for telepathy. However, out of a
grand total of 10,000 calls, 294 were correct as compared
with an expectation of 250, and it can be shown that the
odds against this happening by chance are roughly of the
order of 160 to 1. Odds of 20 to 1 are usually considered
significant in psychological work, and odds of 100 to 1 very
significant. If this had been any ordinary kind of psycho-
logical experiment, the experimenter would undoubtedly
have concluded that the experiment had been successful in
proving the existence of extra-sensory perception. The re-
sults certainly do not bear out Goover's contention that
there is no evidence for telepathy or clairvoyance in his
work, but, on the other hand, we cannot either accept it as
positive proof. The main weakness is an experimental one.
As S. G. Soal and F. Bateman have pointed out in Modern
Experiments in Telepathy, Coover's method of obtaining a
series of call cards that was approximately random by means
of hand-shuffling a solitary pack of cards was not very
efficient. A pack of cards tends to cut more easily at certain
places than at others, and if the card situated in such a place
happened to be a popular one, such as the nine of diamonds
or the ace of spades, this card would turn up more frequently
than the others, and would also be guessed more frequently,
thus increasing the number of correct guesses.

As far as it goes, then, Coover's pioneer experiment, while

Telepathy and Clairvoyance 123

technically faulty, did produce evidence in favour of extra-
sensory perception, the odds being something like 160 to 1.
Strangely enough, he is still often quoted as having 'dis-
proved' the existence of extra-sensory perception. This is an
interesting case of wishful thinking and suppression of evi-
dence, and unfortunately many other such cases can be
found in the literature of the subject.

A rather different type of experimental set-up was used
by three Dutch investigators, Heymans, Brugmans, and
Wynberg. The guesser sat blindfolded by himself in front of
a large chess-board containing forty-eight squares instead of
the usual sixty-four. There were six rows numbered 1 to 6
and eight columns with the letters A to H, so that any of the
forty-eight squares could be located by the letter and the
number, as in the Continental system of chess notation. The
experimenters were in a room immediately above that in
which the guesser was seated, and watched him through a
thick pane of glass let into the floor. They selected one of the
forty-eight squares by drawing a card from each of two
shuffled packs, one containing the numbers 1 to 6 and the
other the letters A to H. The subject was then observed to
make his choice, and the choice was written down and com-
pared with the correct position. Their best subject, a young
man called van Dam, was right sixty times in 187 trials, as
compared with a chance expectation of about four suc-
cesses. This, again, is far beyond any reasonable level of
statistical significance for the acceptance of chance as a
likely explanation, but, again, there are certain criticisms of
the experimental method. Conditions of observations were
unfavourable for the experimenters, and they may well have
made mistakes by counting near successes as successes.
Again, the observers, who knew the correct square, watched
the subject while he was groping over the board, and it is
not entirely impossible that they became excited when his
hand was in the region of the right square and by some
slight movement they might have produced auditory cues
for him to pick up. The reader may consider these objections
as trifling and unlikely, but there is ample evidence in the

124 Pari One ' Borderlands of Knowledge

history of the subject that a very small auditory and visual
cue can provide a suitably acute subject with all the in-
formation needed.

Two examples may be quoted. The first one is the case of
the Elberfeld horses. These horses, which have enjoyed a
considerable degree of renown, worked out the answers to
arithmetical problems given to them by spectators, and ap-
parently showed an uncommonly high degree of intelligence,
even for horses. Although many investigators were con-
vinced that the phenomena produced were genuine, it was
finally established that the horses were responding to slight
movements made by their trainer, probably quite uncon-
sciously. Many stage magicians who find hidden objects rely
on unconscious movements made by the spectators who
know the place where the object is hidden, and by respond-
ing to these movements succeed in their search.

A classic example in the use of subliminal hearing cues is
the case of a Latvian peasant child, Ilga, who was supposed
to be able to read by telepathy any book that her mother
was looking at. The mother was always in the background,
encouraging her, while the girl read falteringly almost
syllable by syllable. By taking sound recordings of the
mother's interjections of encouragement, and by playing
these back to Ilgar, it was discovered that the girl was re-
sponding to very slight auditory cues picked up from her
mother. These cues were not noticed by quite a large num-
ber of investigators, who were taken in by Ilgar's per-
formance.

Again, some performers make use of signal cues, as in the
case of the blindfolded 'clairvoyant' who sits on the stage
while his confederate goes round the auditorium asking for
articles of various kinds to be handed to him, which his col-
league then guesses. These guesses are based on various types
of auditory cues elaborated into a system by the performers,
which completely escape the notice of the spectators. There
is some evidence that quite a few successful bridge partner-
ships have been based on the same principle !

It might be argued that Heymans and the other experi-

Telepathy and Clairvoyance 125

menters were highly competent psychologists and would be
on the look-out for this kind of thing. This, of course, is per-
fectly true, but it is not really a relevant argument. An ex-
periment, to be acceptable as decisive in a matter of this
kind, must not depend on one's faith or belief in the profes-
sional competence of anyone. The design must be com-
pletely foolproof in order to be acceptable. If there is the
slightest possibility, however unlikely, that the subject might
have succeeded by means other than extra-sensory ones,
then the experiment fails to prove the case. How necessary
this general rule is will be realized when it is remembered
that both the Elberfeld horses and Ilga were investigated
and accepted as genuine cases of telepathy by scientists of a
standing equal to that of the Dutch investigators, whose
work we are now considering.

The next to investigate telepathy was G. H. Estabrooks,
who used an ordinary pack of playing-cards. He took the
precaution of having sender and receiver and guesser separ-
ated in the two halves of a relatively sound-proof room,
separated by a partition with double doors which were
closed during the experiment. He also carried out long-
distance experiments in which the sender and the guesser
were in two rooms sixty feet distant from each other. His re-
sults were significant at the 1,000 to 1 level, but the number
of trials was relatively small because Estabrooks seems to
have become discouraged by the fact that subjects who start
out by making positive scores later on make negative ones.
This has been found in many investigations since and ap-
pears to be an almost universal psychological fact. It is a
pity that Estabrooks did not continue his work because,
technically, it appears to be the best until it was superseded
by more recent investigations, and the results, while promis-
ing, do not reach a level of significance which would make
the acceptance of extra-sensory perception mandatory.

We now come to the work carried out by J.B.Rhine at
Duke University in North Carolina. The countless studies
carried out by this investigator, his wife, and numerous
colleagues and students, have probably done more than

126 Part One • Borderlands of Knowledge

anybody else's to put parapsychology 'on the map'. At the
same time, Rhine's experiments are so well known that any
detailed recounting of them would not have very much point.
I shall, therefore, recount them in less detail than their im-
portance would obviously deserve.

Rhine's main innovation in technique was the introduc-
tion of a new type of experimental material. Where pre-
viously playing-cards and numbers had been used, he now
introduced the so-called Zener cards, which had inscribed on
them one of five symbols - a circle, square, five-pointed star,
wavy lines, or plus sign. The cards were made up into packs
of twenty-five, each pack containing five cards of each symbol.

This may seem a very slight change, but for many reasons
it is a very important one. Playing-cards and numbers, like
human beings, have their favourites and their wall-flowers.
Among cards, the ace of spades and the nine of diamonds are
much more likely to be called than the four of clubs or the
ace of hearts. Among numbers, seven is much more likely to
be called than two. Playing-cards and numbers, therefore,
have associated values which interfere with the calculation
of chance agreements, and which upset an experiment based
on the equivalence in the minds of the subjects of all the
symbols which are available. In choosing five relatively
innocuous symbols, as Rhine had done, most of these diffi-
culties were eliminated, although even this method does not
completely get rid of them.

Rhine tried out various techniques with these cards in an
attempt to separate out telepathy from clairvoyance. Some-
times in telepathy tests the sender would look at each card in
turn and ' transmit ' that to the receiver and then put them
down. Sometimes in clairvoyant tests the sender would just
lift each card in turn without looking at it, and then put it
down again, to be checked later against the receiver's
guesses. In the so-called 'down-through' technique, the
sender would just look at a pack of Zener cards and the re-
ceiver would try to guess the sequence of cards without their
being touched at all.

With these various conditions, Rhine reported astonishing

Telepathy and Clairvoyance 127

successes which set the odds against chance soaring into the
millions. However, the results were much criticized on
various grounds. Some of these are statistical and are related
to the fact that each pack contained exactly five cards bear-
ing each of the five symbols. This is not strictly a chance
arrangement and requires a slight modification in the usual
formula for calculating probabilities. The correction re-
quired is very small, however, and could not possibly ac-
count for Rhine's results. In later work the correct formula
has been used, or else the sequence of the symbols has been
established on the basis of a table of random numbers, a pro-
cedure which is perhaps preferable. In any case, the statis-
tical criticisms which have been brought forward against
Rhine, with this one exception, are not valid, and there is no
doubt that if modern experiments on telepathy are vulner-
able to criticism, it must be on the grounds of experimental
rather than of statistical inadequacies. A lot of nonsense is
talked on this point, even by people who should know bet-
ter, and it should be emphasized that the highest authorities
on mathematical statistics have explicitly given their bless-
ings to the methods of analysis currently used.

Another criticism made is that of inadequate shuffling.
As most card-players know, the ordinary process of shuffling
and cutting does not destroy completely sequences of cards,
and if in telepathy experiments the habits of the guesser
happen to fit in with the sequence of cards, then it is pos-
sible that these sequences may persist through several
processes of shuffling, thus producing slightly inflated
scores. While it is perfectly true that successive arrange-
ments obtained by repeated shufflings of a pack of Zener
cards are not perfectly random cards drawn from the
623,360,743,125,120 possible arrangements of the twenty-
five cards, yet there is ample experimental evidence that
high scores made by good subjects cannot be accounted for
by such differences. The method of proof, which is a very
simple one, is that of matching guessing sequences of a given
person against runs of cards for which they were not in-
tended. Thus we might compare the set of guesses made by

128 Part One • Borderlands of Knowledge

Mr Smith with the set of cards shown on another occasion
to Miss Doolittle, and the set of guesses made by Miss
Doolittle with the run of cards but before Mr Smith. In
half a million such matchings the average score was found
to be 4*9743 as opposed to the chance value of 5-00. Thus,
this criticism cannot account for the average scores of 7, 8,
or more scored by successful subjects.

Optional stopping has been another alleged cause of high
scores. As by chance people's scores will be equal often
above and below the chance level of 5 successes per 25 calls,
it is suggested that a high score can be built up by simply
going on with each subject until he reaches a point where he
has an above-average score. The run is then stopped before
the subject can go down into his next negative swing. By
accumulating a large number of subjects and following the
same procedure with each, we would then arrive at the large
number of positive deviations which, when added together,
will give high odds against chance. This is an interesting
and, at first sight, plausible theory, which in a slightly
adapted form has led many people to lose millions of francs
at the roulette wheels of Monte Carlo. As Soal and Bateman
point out the main difficulty with the argument is this.
Some people start out with a negative score and go on hav-
ing negative scores on the average. To convert these into
positive scores requires so many runs that the over-all
significance of the total results is reduced to practically zero
because of the large number of runs required to produce the
positive total. It is possible, by means of optional stopping,
to increase the chance score slightly above 5, but this can be
taken into account by suitable statistical formulas and it
cannot, under any circumstances, produce the persistent
high scoring observed in some subjects.

Unconscious whispering has also been suggested as a
possible explanation of some of the findings, and we have
already seen that this cannot be ruled out too quickly.
However, experiments have been conducted with a distance
of several miles between sender and recipient, and it can
hardly be suggested that an unconscious, or subliminal,

Telepathy and Clairvoyance 129

whisper would carry that far ! In the clairvoyant tests the sen-
der himself had no notion of the sequence of cards, and thus
could hardly give them away by whispering. The same is true
of the calls made by means of the down-through technique.

Errors in recording and checking have been considered as
possibly accounting for some of the extra-chance results, the
hypothesis being that the investigator would tend to make
errors in favour of his preconceived notion, counting as co-
incidences guesses where, in fact, card and guess were differ-
ent. This again is not a likely explanation. Repeated
checking has shown that errors are almost negligible and
that where they do occur they tend much more frequently
to indicate a correct matching missed rather than an incor-
rect matching scored as correct. Considerable care is taken
in modern experiments to make duplicates of all card and
guess sequences, and these are usually independently
checked by several people.

Improper selection of data is the last of the main critic-
isms brought against Rhine. It is suggested that perhaps
groups of data giving only chance results were rejected and
only those giving positive results reported. As an alternative,
it is suggested that perhaps the positive results reported by
some investigators are counterbalanced by the more numer-
ous negative results which may have been obtained by
others, but not reported because of their negative nature.
The first of these arguments impugns the integrity of the in-
vestigator, and there is no evidence that Rhine fell into this
error. The fact that many other previously hostile investiga-
tors who were fully aware of this difficulty confirmed his
positive results tends to disarm this line of criticism. As re-
gards the second version of this argument, it can be shown
that it would require quite impossibly large series of negative
instances to reduce the obtained positive results to insigni-
ficance. This argument too, therefore, must be rejected.

One further criticism must be mentioned because much
has been made of this. It is suggested that certain irregulari-
ties on the backs of the cards may have given the subjects
certain cues; that occasionally the symbols may have been

130 Part One • Borderlands of Knowledge

printed so heavily that they could be seen through the back
of the cards; and that it was not always impossible for the
subjects to see directly the bottom card of the pack, or the
card where the cut was made. These possibilities cannot be
ruled out for some of the early Rhine experiments, but they
certainly do not obtain in most of this work. The cards were
usually screened from the receiver, or the receiver and sen-
der were in different rooms. Usually the subject did not
know when he had scored a success and when not, so that
he had no opportunity for associating the back of the card
with the symbol on the face of it.

It is important to realize in considering the storm of critic-
ism that broke over Rhine's head that, at least in part, he
was responsible for much of it. There are, roughly speaking,
two ways of carrying out work in this field. One method -
namely, that adopted by Rhine - is to start out with very
unstandardized and loose conditions which give the subject
a more positive, co-operative type of attitude. This attitude,
apparently, is an important psychological factor in success-
ful work, and it tends to be destroyed by the imposition of
stringent controls. Gradually, Rhine would tighten up his
controls, always trying to retain the positive attitude on the
part of the subject, until finally he would reach a point
where controls were completely stringent and where yet
positive scores were still obtained from the subject. An
alternative method would have been to start right away with
the more stringent controls, but it is quite possible that
under those conditions co-operation would have suffered
and less significant scores would have been obtained.

Rhine's method is probably the superior one, but it is
doubtful whether he was right in publishing the results from
his less rigidly controlled experiments. The criticisms rightly
levelled against these have prejudiced many people against
accepting the more rigidly controlled experimental investi-
gations which followed, and it would probably have been
better to have restricted the account only to a small number
of cast-iron experiments which left no loophole at all.

One of these highly impressive experiments was carried

Telepathy and Clairvoyance 131

out on Hubert Pearce, one of Rhine's star subjects. Subject
and sender were in different buildings at Duke University
while the experiment was being carried on, there being no
telephonic communication between the two buildings. Ac-
cording to a prearranged timing schedule, Dr Pratt, who
carried out the experiment, lifted out the top card from the
shuffled pack at a specified time and laid it face downwards
on the centre of the table. Thirty seconds later, Pearce wrote
down his guess, and after another thirty seconds, Pratt lifted
off the next card. Thus, they worked at the rate of one card
a minute, with fifty guesses at each sitting. Pratt and Pearce
would then seal up their record sheets and deliver them in-
dependently to Rhine, who checked the results. An average
over 750 trials of 8-7 hits per 25 cards was scored, giving
odds against chance of something like 1 o20 to 1 . The reader
may like to try out the various criticisms made and evalu-
ated above and see for himself how little they can account
for the effects observed in this study.

During? the years in which Rhine tightened up his con-
trols there seems to have come about a definite change in
outlook. At first, it used to be believed that almost anyone
had parapsychological powers, if only to a slight extent, and
that experiments involving large numbers of people would,
on the whole, tend to give positive results. It is not incon-
ceivable that the positive results achieved under those con-
ditions were, wholly or in part, due to laxity of experimental
conditions. In more recent years attention has shifted rather
to a small number of highly gifted individuals who can be
relied on to give high scores over long periods. Large-scale
experiments involving many people tend to give only chance
results, although they may be useful in discovering the rare
gifted person who can then be studied further.

It would not be very interesting to recount in detail the
many confirmatory experiments carried out by Rhine and
by others who have followed in his footsteps over the years.
Unless there is a gigantic conspiracy involving some thirty
University departments all over the world, and several hun-
dred highly respected scientists in various fields, many of

132 Part One • Borderlands of Knowledge

them originally hostile to the claims of the psychical re-
searchers, the only conclusion the unbiased observer can
come to must be that there does exist a small number of
people who obtain knowledge existing either in other
people's minds, or in the outer world, by means as yet un-
known to science. This should not be interpreted as giving
any support to such notions as survival after death, philo-
sophical idealism, or anything else; the interpretation of
this fact must await a more thorough knowledge of the con-
ditions under which such results can be obtained and the
means through which extra-sensory knowledge is trans-
mitted.

Rhine and his associates tried out various methods to dis-
cover the correlates of this extra-sensory ability. In particu-
lar, they were concerned with the question of which kind of
personality was most likely to give positive E.S.P. results.
Intelligence, although the most obvious candidate, was soon
ruled out because there appears to be no good cause to sus-
pect extra-sensory perception to occur more frequently
among people of high, medium, or low intelligence. Atten-
tion was therefore directed towards the non-cognitive side
of personality. One of the first to work along these lines was
Humphrey, who made use of a drawing test. This test,
originated by Elkisch, makes use of the types of drawings
produced by experimental subjects, contrasting the ' expan-
sive type', which shows imagination, vitality, and freedom
of expression, with the 'compressive type', which shows in-
hibition, lack of imagination, and timidity.

Humphrey got her subjects to make a sketch aimed at a
drawing contained in an opaque sealed envelope in the hope
that there would be some relation between the expansive-
compressive quality of the sketch and the clairvoyant ability
of the person to respond to the drawing in the envelope. A
complex scoring system was elaborated, and Humphrey
found that the 'expansives' scored on the average above
chance expectation, while the 'compressives' scored below
the chance level, the difference being highly significant with
the odds against chance in the neighbourhood of 300,000 to

Telepathy and Clairvoyance 133

one. With the total score of the expansives showing a posi-
tive deviation, and the total score of the compressives show-
ing a negative deviation, the combined total score did not
significantly deviate from chance, the total score thus show-
ing no evidence of extra-sensory perception.

In a further experiment telepathy was used rather than
clairvoyance by having the sender look at the pictures in a
distant room, and the guessers drawing sketches in another
room. This time the relationship was reversed, the compres-
sives scoring above-chance level and expansives scoring be-
low. In conformity with this finding, another investigator
found that expansive subjects, selected because they had
produced high scores in clairvoyance tests, scored signi-
ficantly below chance expectation when they were made to
do telepathy tests. Conversely, compressive subjects who had
scored below chance expectation in clairvoyance tests ob-
tained large above-chance scores with telepathy. Repetition
of these experiments in Great Britain has not succeeded in
duplicating the results reported from America, and it should
be noted in any case that the Elkisch technique is not a very
reliable one, as one and the same person may make com-
pressive drawings on one occasion and expansive ones on
another. Altogether, it is probably a safe rule here, as in
other fields of psychology, to insist that a finding, before it
can be accepted, should be duplicated in several different
departments. Until this is done we can hardly accept Hum-
phrey's results as being anything but suggestive.

Much the same is true of experiments carried out by
Schmeidler. She divided her subjects into two groups, whom
she called sheep and goats respectively. The goats had ex-
pressed the conviction that telepathy was impossible under
the conditions of the experiments, while the sheep con-
sidered that it might occur. On the average, the sheep
tended to score little above five and the goats a little below.
On 185,725 guesses the average scores were 5*15 and 4/92
respectively.

Next Schmeidler went on to use the Rorschach ink-blot
test on her subjects as a measure of emotional adjustment or

134 Pari One ' Borderlands of Knowledge

stability. She then divided her subjects into four groups:
well adjusted and poorly adjusted sheep, and well adjusted
and poorly adjusted goats. Her hunch was that well adjusted
sheep would get higher scores than poorly adjusted sheep,
and that well-adjusted goats would get larger negative de-
viations than poorly-adjusted goats. Thus, in her view, poor
adjustment might inhibit both the positive and the negative
scoring ability of subjects. On the whole, her surmise ap-
pears to have worked out, but the differences are very small
throughout. In addition, it should be noted that in her cal-
culations she did not employ the correct formula, although
the results would probably still have been significant if she
had done so.

For various reasons, these results are difficult to accept.
We know so little about telepathy and clairvoyance that one
might be inclined to say that anything is possible and that
these results are no more miraculous than others which we
have considered previously. However, we do know some-
thing about the Rorschach test, and, as we shall see in a later
chapter, this test is very unreliable and almost completely
lacking in validity as far as the measurement of good or poor
adjustment and emotional stability are concerned. Under
those conditions, it seems exceedingly unlikely that such a
test should succeed in picking out good and poor subjects re-
spectively. Until these results are repeated we certainly
should be extremely chary of accepting the implications con-
tained therein. To show how necessary such scepticism is,
we may perhaps recall that the particular version of the
Rorschach used by Schmeidler had early on been used by
another investigator to predict success at college with con-
siderable accuracy. When the study was repeated else-
where, however, results never went beyond chance. Repeti-
tion of experiments is our only safeguard against accepting
false conclusions. On the whole, therefore, we must say that
nothing is definitely known with respect to the personality
traits which correlate with parapsychological ability, al-
though a few promising leads may be contained in some of
the work briefly summarized above.

Telepathy and Clairvoyance 1 35

We must now turn to quite a different aspect of telepathy
and clairvoyance, which in a sense is much more miracu-
lous and hard to accept than anything that has been men-
tioned so far. Throughout the older literature there is appar-
ent a belief in the existence of precognition, i.e. the ability to
look into the future and predict what will happen. Dreams
which foretell the future have often been reported and will
be considered in our next chapter, but the experimental in-
vestigation of prediction and precognition had to await the
experimental work of Rhine, Tyrrell, and Soal. In Rhine's
work what essentially happened was that the receiver called
out the sequence of cards in a Zener pack, before that pack
had been shuffled; these calls were then matched with the
subsequent order of cards. Scoring was above chance, but
only very little so, and as certain alternative hypotheses may
account for his findings they are not considered completely
convincing by certain critics.

Tyrrell's work is noteworthy because it is based on a ma-
chine which effectively gets rid of the human element in
shuffling, recording, and so forth. This apparatus consisted
of a row of five boxes, each of which contained a small elec-
tric lamp. The lamps were connected by wires to five keys
which were operated by the experimenter from a desk situ-
ated a few feet away. The lids of these boxes could be lifted
by the subject, who sat behind a large screen. When a key
was pressed by the experimenter, the corresponding lamp
was lighted, and the subject, at a signal from an electric
buzzer, lifted the lid of what she thought to be the correct
box. (The lamp was not allowed to light up before the lid
was lifted, as otherwise the subject might have obtained
slight cues from the heat of the box or the light which might
have emanated from it.) The opening of any box automatic-
ally drew a line upon a roll of paper, and success was re-
corded by a double line. A commutator in the circuit
changed the connexions between lamps and keys all the
time, so that the experimenter did not know which lamp he
was lighting when he pressed a given key. Alternatively, the
keys could be dispensed with and a rotating switch with a

136 Part One • Borderlands of Knowledge

single arm brought into play to act as a mechanical selector.

Under all these conditions, telepathy was ruled out, as
Tyrrell himself would not know which was the correct box.
Tyrrell's main subject, a Miss Johnson, was extremely suc-
cessful in trials of clairvoyance, but she also succeeded in
trials of precognitive clairvoyance, which makes her suc-
cesses very much more interesting. In these trials, Miss
Johnson opened a box about half a second before the experi-
menter pressed the key. There were 2,255 such trials with
539 hits, which is in excess of 88 above chance expectation,
with odds against chance of about 270,000 to one. Inspection
of the automatic recording on the paper showed that in the
case of every hit the box had really been opened before the
key was pressed. It would be very difficult indeed to explain
these results away.

Most impressive perhaps of all the studies in precognition,
however, is the work of S.G.Soal, a British mathematician
who started out as a sceptic and for many years failed to
produce positive results. Encouraged by another investiga-
tor who claimed to have found evidence of precognition,
Soal went back over a large portion of his data to seek for
what are now called 'displacement' effects. According to
the theory, a person aiming at the card which is his target
for the moment may actually miss it and displace his guess
to the card immediately preceding or immediately following
the one shown. Soal found, indeed, that two of his subjects,
Basil Shackleton and Gloria Stewart, had given highly signi-
ficant scores on the cards, one behind and one ahead of the
target at which they were aiming. Shackleton in particular
was found in subsequent investigations persistently to dis-
place his aim to the card one ahead. Shackleton could, when
specially pressed, shelve his guessing to the target card, but
found it more congenial and natural to guess the card not
yet exposed. When the rate of calling the cards was speeded
up, Shackleton's guess would shift to a card two ahead so as
to preserve his usual time relation.

These experiments do not necessarily prove the existence
of precognition. The target sequence was predetermined by

Telepathy and Clairvoyance 137

a list of random numbers, and if the subject could obtain
knowledge of this list by telepathy he could then deduce the
card one or two ahead without using precognition. How-
ever, Soal tested this hypothesis by arranging for the target
order at certain sessions to be determined on the spot by
drawing counters from a bowl. An assistant would pick up a
counter and show it to the sender, who would instantly look
at the card corresponding to the number shown on the
counter. Still the pre-intuitive effect continued at much the
same level of significance.

Taking together the work of Rhine, Tyrrell, Soal, and
other investigators, such as Carington, who used the tech-
nique of getting people to draw pictures of objects which he
was trying to 'send', there appears to be little possibility of
denying the existence of precognition. However much such
a conclusion may go against our ingrained habits of thought,
the experimental rigour and the statistical adequacy of the
experiments are such as to make criticism impotent. Unless,
again, all these people and the many independent scorers
and colleagues involved in this work are actively fraudulent,
the conclusion is inescapable that certain people possess a
faculty which enables them to foretell events in the imme-
diate future.

Displacement effect is one of many secondary effects
found in parapsychological research. Other such secondary
effects have been known for quite a long time. Thus, it is
known that many subjects tend to show a decline in their
ability as time goes on; few subjects retain their abilities
over more than two or three years. Even on a given occasion
there tends to be a decline from the calls made in the first
ten minutes of a three-hour session, say, to those made to-
wards the end of that session. It is sometimes possible to de-
press the rate of calling below a chance level, so that people
who start out giving highly positive scores decline until they
give significantly negative ones at the end. If this happened
just once or twice it might, of course, be just a chance effect,
but it has been found to happen consistently with many sub-
jects and in many different experiments. Another secondary

138 Part One • Borderlands of Knowledge

effect has been labelled salience and relates to the observa-
tion that within a given run of twenty-five cards, most hits
will be made at the beginning rather than in the middle of
the series. The fact that secondary effects of this type have
been found in many different investigations suggests that
certain general laws may yet be found in this complex field
and that we may eventually be led to a better understanding
of these events.

We have mentioned the difficulty of discriminating be-
tween telepathy and clairvoyance. Our brief discussion of
precognition will indicate just why the experimental task of
differentiation is so enormously complicated. Suppose that
our procedure was something like this. The experimenter
opens a new pack of playing-cards in a dark room, puts
them through a mechanical shuffler, and finally selects the
target cards. The receiver in another dark room guesses
these cards. Yet it cannot be said that clairvoyance has
taken place, because someone will have to check the guesses
against the order of the actual cards, and will thus have to
perceive these cards in the normal way. The person who car-
ried out the guessing correctly might have obtained his
knowledge through clairvoyance, but he might also have
been in contact with the person who carried out the check-
ing by means of telepathy of a precognitive type. Thus, this
experiment would not be adequate for establishing clair-
voyance as opposed to telepathy. One possible way out of
the dilemma has already been mentioned ; it is the mechani-
cal scrambler and recording device by Tyrrell.

Similar difficulties arise in trying to prove the existence of
telepathy. Suppose the experimenter thinks of the names of
various Zener cards without having any actual physical
cards in front of him, while the receiver writes down his
guesses in a different room. If now the experimenter writes
down the names of the symbols he had thought of in their
proper order, and a comparison with the receiver's guesses
reveals evidence of E.S.P., we can still not sav that we have
proved the existence of telepathy because of the possibility
of precognitive clairvoyance on the part of the receiver of

Telepathy and Clairvoyance 139

the future physical record made by the experimenter. The
way out of this difficulty is by means of a code known only
to the experimenter, and never written down by him, by
means of which he translates numbers drawn at random
from an urn into symbols which he then tries to transmit to
the receiver. The numbers are written down but not the
code which translates them into the symbols, so that while
pre-intuitive clairvoyance might enable the receiver to know
the numbers drawn, this would still not be sufficient to en-
able him to translate these into the symbols he is actually
asked to guess.

Experiments on both these lines have shown that both
clairvoyance and telepathy must be presumed to exist. It has
also been found that while many people who possess the one
type of ability also tend to possess the other, this is by no
means necessary, and in certain cases individuals may obtain
high scores by telepathy, but not by clairvoyance and vice
versa. The great majority of experiments, however, do not
attempt to differentiate between these two abilities and must
simply be taken as evidence of some all round parapsycho-
logical or extra-sensory perception capacity on the part of
successful individuals.

A few words may be said in conclusion about psycho-
kinesis, the last of the many parapsychological phenomena
experimentally investigated. Rhine's method was simply
one of letting dice roll down an inclined plane and getting
his subjects to will the dice to go down in such a way as to
give either a high (6, 5) or a low (1, 2) score. Routine
methods of rolling the dice were developed so as to make
them independent of the subject, and it has been found that
strongly extra-chance scores could be obtained in this way.
Certain experimental difficulties are obvious from the first;
thus, dice tend to fall in such a way that the 6 or the 5 is
exposed more frequently than any other number because
the slight reduction in weight of those faces having a large
number of spots hollowed out from the cube makes these
sides come up more frequently. However, by getting his sub-
jects to call whole series in an attempt to get either high or

140 Part One • Borderlands of Knowledge

low scores, Rhine has successfully overcome this difficulty.
We should be cautious in accepting these results until they
have been duplicated successfully elsewhere, and altogether
the amount of work done on this phenomenon is tiny com-
pared with that carried out in the field of telepathy and
clairvoyance. The evidence certainly is impressive, but not
in my view conclusive, and it will need a few more years be-
fore any definite conclusions will be possible on this very
important question.

Ideally, at the end of a chapter such as this, there should
appear a few paragraphs dealing with theories to explain the
observed phenomena. While many such theories have been
suggested, there is little point in going into them in detail
because they cannot be regarded as throwing very much
light on our subject. Before theories can be very fruitful in
science a large number of facts require to be known. Little is
known in the field of extra-sensory perception, except that
something seems to be active there which cannot be ac-
counted for in terms of our ordinary rules and theories. In
view of the difficulty of isolating the phenomenon and of
finding good subjects, experimentation has not essentially
gone very far beyond this. Certain rather interesting
secondary facts are known, but much more progress will
need to be made in their specific delineation before the
search for a reasonable theory will become practicable. This
is a disappointing conclusion, and it undoubtedly accounts
for some of the hostility which orthodox science has shown
towards this phenomenon. Isolated facts are not easily
woven into the fabric of science, and as yet there is not in
parapsychology a consistent system of facts linked together
by hypotheses and theories. Some slight beginnings there
are, as in the work linking personality and parapsychology,
but, as we have seen there, these links are very weak and
hardly to be trusted.

One of the reasons for this state of affairs, of course, is to
be found in the relative youth and immaturity of para-
psychology; another is the difficulty of obtaining the neces-
sary finance for carrying out concentrated, co-ordinated,

Telepathy and Clairvoyance 141

long-term work of this kind. A third reason perhaps can be
found in the hostility which, until recently, research in para-
psychology has aroused, and which is even now far from be-
ing completely overcome. In time, no doubt, all these diffi-
culties will vanish and we will have the beginnings of a
rational theory of parapsychology. Until then, we should
perhaps not be too critical and remember that we have no
proper theory of most psychological phenomena either, a
fact which does not prevent us from recognizing that
psychological phenomena do occur and are worthy of
scientific investigation.

4

THE INTERPRETATION OF DREAMS

Mankind has always been interested in dreams, and many
attempts have been made to interpret the meaning of them.
The reasons for this interest are not difficult to find. Dreams
are odd and striking phenomena, similar to waking thought
in some ways, but quite dissimilar in others. The objects
which enter into the dream are usually everyday kinds of
objects, like horses and trains and people; similarly, the
place where the dream occurs is usually a familiar one, like
the dreamer's house or a field or a night club. Yet what hap-
pens in the dream is often quite unlike the happenings of
everyday life. People may change into each other or into
animals ; the dreamer may be wafted through the centuries
or across the oceans in no time at all, and the most frightful
and wonderful events may happen to him - more frightful
and wonderful than any described in The Thousand and One
Nights. In addition, dreams often have a very strong emo-
tional content. This is obvious enough in the case of the
nightmare, but even in the case of more ordinary dreams,
strong emotions, both pleasant and more usually unplea-
sant, may be called forth. The presence of such emotions in-
creases interest in dreams and makes people feel that they
must be of some importance.

Factual information is scant and not very illuminating.
When large numbers of dreams of people in our culture are
examined, we can make a rough statistical analysis of the
settings in which the dreams occur, the characters appearing
in them, the actions through which they go, and the emo-
tions which they betray. Most dreams have some fairly de-
finite setting. In only about 5 per cent is the dreamer un-
aware of the setting. In another 15 per cent the dreamer is
in a conveyance such as an automobile, a train, an aero-
plane, a boat, an underground, or a tram. Roughly 10 per
cent of dreams are set in recreational surroundings : amuse-

142

The Interpretation of Dreams 143

ment parks, at dances and parties, on the beach, watching
sports events, and so on. More frequent than any of these
settings, however, is the house or rooms in a house; these
account for some 35 per cent. Apparently the living-room is -
the most popular, followed in turn by bedroom, kitchen,
stairway, basement, bathroom, dining-room, and hall. An-
other 10 per cent of dreams are set in rural and out-of-doors
surroundings. Men's dreams tend to occur more frequently
in out-of-door surroundings, women's more frequently in-
doors. In another 10 per cent of dreams the dreamer is walk-
ing along a street or road. The remaining dreams are diffi-
cult to classify with respect to their settings.

Psychoanalysts often try to interpret certain aspects of
the dream by taking into account the setting. The fact that
the dream occurs in a conveyance, for instance, is inter-'
preted in terms of the fact that the dreamer is going some-
where, is on the move; movement represents ideas such as
ambition, fleeing from something, progress and achievement,
breaking family ties, and so forth. Trains, automobiles, and
other vehicles are instruments of power, and are thus inter-
preted as symbols for the vital energy of one's instinctual im-
pulses, particularly those of sex. Recreational settings are
usually sensual in character, being concerned with pleasure
and fun, and imply an orientation towards pleasure rather
than work. A symbolic interpretation of this kind may be
even more highly specialized; thus a basement is supposed
to be a place where base deeds are committed, or it may
represent base unconscious impulses. We shall be concerned
with the validity of such interpretations later on; here let us
merely note the fact that interpretations of this type are
made by some people.

In addition to a setting, the dream must also have a cast.
In about 15 per cent of all dreams no one appears but the
dreamer. In the remaining 85 per cent, on the average an-
other two characters appear. Most of these additional char-
acters are members of the dreamer's family, friends, and
acquaintances. About 40 per cent of the characters in our
dreams are strangers; they are supposed to represent the

144 Part One ' -Borderlands of Knowledge

unknown, the ambiguous, and the uncertain ; sometimes they
are interpreted as alien parts of our own personality which
we may be reluctant to acknowledge as belonging to us.
Prominent people are seldom found in dreams ; this may be
because our dreams are concerned with matters that are
emotionally relevant to us.

What does the dreamer do in his dream? In some 35 per
cent of the cases he is engaged in some kind of movement,
such as walking, driving, running, falling, or climbing.
Mostly these changes in location occur in his home environ-
ment. In another 25 per cent of dreams passive activities
such as standing, watching, looking, and talking are in-
dulged in. There appears to be an absence of strenuous or
routine activities in dreams - there is little in the way of
working, buying or selling, typing, sewing, washing the
dishes, and so forth. When energy is being expended in the
dream it is in the service of pleasure, not in the routine
duties of life. Women, generally speaking, have far fewer
active dreams than men.

All kinds of emotions are attached to the actions and persons
making up the dream, as well as to the settings. Quite gener-
ally unpleasant dreams are more numerous than pleasant
ones, and apparently as one gets older the proportion of un-
pleasant dreams increases. The unpleasant emotions of fear,
anger, and sadness are reported twice as frequently as the
pleasant emotions of joy and happiness. Emotion in dreams is
often taken to be an important aid in interpreting the dream,
as we shall see later. In this it differs very much from colour;
about one dream in three is coloured, but the attempt to
find any kind of interpretation whatsoever for the differ-
ence between coloured and black-and-white dreams has
proved very disappointing.

This, then, is the kind of stuff dreams are made of. What
kind of interpretation can we make of them, assuming that
they carry some form of meaning, however distorted, and
are not merely the random after-effects of sensory stimula-
tion encountered during the day or the immediate accom-
paniment of sensory stimulation present at the time the

The Interpretation of Dreams 145

dream is taking place? Both these theories have been held,
and may account for some dreams at least. We have seen a
beautiful car during the day, and it may appear in our
dreams. The alarm clock may strike, and instead of waking
up we may have a dream of a church clock calling the de-
vout to prayer.

A rather more adequate theory, and certainly a more
poetic one, is held by savage tribes who believe that when a
person is asleep his soul flies away and experiences in actual
fact those things which to the sleeper appear as a dream.
Such a view of course leads to complications. Thus a certain
tribal chief, on hearing that one of his subjects had dreamed
of having intercourse with one of the chief's daughters, de-
manded of him the full bride's money, on the theory that the
sleeper's soul had in fact obtained the satisfaction which
could in law be obtained only on paying the chief the price
of his daughter. The reader may like to consider the ethical
implications of this theory and ponder over possible experi-
mental methods of disproving it; we shall, instead, sum-
marily dismiss it, as we did the previous theory.

For the great majority of dreams such explanations as we
have considered are clearly insufficient, and we encounter
next two great groups of theories which attempt to interpret
and explain dreams.

According to the first of these theories, dreams are pro-
phetic in nature ; they warn us of dangers to be encountered
in the future, they tell us what will happen if we do this
or -that; they are looked upon as guide-posts which we may
heed or neglect as we wish. This is probably the most com-
mon view of dreams which has been held by mankind. The
prototype of the prophetic dream is of course Pharaoh's of
the seven fat cattle and the seven lean cattle which Joseph
interpreted. Here we have all the elements of the pro-
phetic dream - the strong emotion indicating the import-
ance of the dream, the symbolic way in which the informa-
tion is wrapped up, and the special skill of the interpreter
who can unravel the secret of the dream and can lay bare its
innermost meaning.

146 Part One • Borderlands of Knowledge

If we take this hypothesis at all seriously, then a study of
the art of dream interpretation clearly becomes of the great-
est possible importance. We need only consider the possi-
bilities opened up by dreaming in advance of the winner of
next year's Derby, to indicate why people have always been
fascinated by this aspect of dream study and why books on
dream interpretation have so frequently been best sellers.
The pattern was set by an Italian scholar called Artemi-
dorus, who lived in the second century of the Christian era.
His book was called Oneirocritics, which means The Art of
Interpreting Dreams, and apart from being translated into
many languages it has been imitated and copied by a very
large number of writers. Essentially, books of this nature are
based on the view that the dream is a kind of secret lan-
guage which requires a sort of dictionary before it can be
understood. This dictionary is provided by the writer of the
dream book in the form of an alphabetical list of things
which might appear in the dream, each of which is followed
by an explanation of its meaning. Thus, if the dreamer
dreams about going on a journey, he looks up 'Journey' in
his dream book and finds that it means death. This may of
course be rather disturbing to him, but he may console him-
self by the consideration that it need not necessarily be his
own death which is being foretold in this fashion.

Few people would take this kind of dream interpretation
very seriously; it is obviously analogous to astrology, tea-cup
reading, and palmistry, in its unverified claims and its
generally unlikely theoretical basis. Nevertheless, some
scientists have taken the possibility of precognition seriously,
as we have seen in a previous chapter, and one or two have
devoted special attention to precognitive powers as shown in
dreams. One of the best known of these is J.W.Dunne,
whose book An Experiment with Time was widely read in the
twenties and thirties of this century.

Let us consider the kind of dream which he considered as
evidence. He reports the occurrence of an unusually vivid
and rather unpleasant dream in which he seemed to be
standing on high ground. This hill or mountain was of a

The Interpretation of Dreams 147

curious white formation with little fissures here and there
from which jets of vapour were spouting upwards. Dunne
recognized the place as an island of which he had dreamed
before; an island which was in imminent peril from a vol-
cano. He remembered reading about Krakatoa, when the
sea, making its way into the heart of the volcano through a
submarine crevice, blew the whole mountain to pieces.
' Good Lord ! ' he gasped, ' the whole thing is going to blow
up.' He was seized with the frantic desire to save the 4,000
unsuspecting inhabitants, and made considerable efforts to
do so by getting the authorities to take some of them off in
ships. Throughout the dream this number 4,000 seemed to
have some special significance.

So much for the dream. A few days later Dunne received
a copy of the Daily Telegraph (he was encamped at that time
near the ruins of Lindley, which was then in the Orange
Free State, and quite cut off from all contact with the
world) . In the paper he read about the explosion of the vol-
cano Mont Pelee, which destroyed the once-prosperous
town of Saint-Pierre, the commercial capital of the French
island of Martinique in the West Indies. Forty thousand
people were said to have lost their lives, and ships were busy
for some time removing survivors.

It will be noted that the number of lives lost was 40,000,
whereas Dunne had dreamed of 4,000, but, as he explains,
when he read the paper he read in his haste that the number
of killed had been 4,000, and in telling the story subse-
quently he always spoke of that printed figure as having
been 4,000. He did not realize that it had really been 40,000
until he copied the paragraph fifteen years later. His ex-
planation of the whole event is that his dream had been
caused precognitively by the reading of the newspaper re-
port, and that consequently the error in the dream respect-
ing the number of dead was caused by the error in reading
the report made subsequent to the dream. Dunne does not
consider the possibility that, on the contrary, his misreading
may have been caused by the memory of the number 4,000
in the dream, although even then we would have to admit a

148 Part One • Borderlands of Knowledge

certain coincidence in his dreaming of a volcano exploding
just before the actual event. Altogether, we can hardly con-
sider this report as very convincing.

Another dream is probably a little more interesting. He
dreamed one night that he was walking down a sort of path-
way between two fields, and separated from them by high iron
railings. His attention was suddenly attracted to a horse in
the field on his left, which had apparently gone mad and
was tearing about in the most frantic fashion. Anxiously he
inspected the railings to see if there was any opening through
which the animal could escape. Finding none, he continued
on his way, only to find, to his dismay, that the animal had
somehow managed to get out after all and was pursuing
him full tilt down the pathway. He ran like a hare in an
attempt to reach a flight of wooden steps rising up from the
path. Next day Dunne went fishing with his brother, when
the latter called his attention to the antics of a horse. There
were two fields with a fence, and a pathway running be-
tween them. The horse was there, behaving just as it had
done in the dream. The wooden steps at the end of the path-
way were there too. There were certain slight differences in
the scene, but by and large they were of no great conse-
quence. Dunne began to tell his brother about the dream,
but broke off because he was becoming worried that the
horse might get out, as it had done in the dream. Failing to
see any gap or even a gate in the railings, he said : 'At any rate
this horse cannot get out', and recommenced fishing; but
suddenly his brother called out to him, and he saw that the
beast had, just as in the dream, got out of the field in some in-
explicable fashion, and was thundering down the path to-
wards the wooden steps. It swerved past these and plunged
into the river, coming straight towards Dunne and his
brother. The end of the story was rather tame, however, for
on emerging from the water, the animal merely looked at
the pair of frightened humans, snorted, and galloped off
down the road. Again, there certainly is a somewhat surpris-
ing coincidence, but it is by no means impossible that Dunne
had previously seen the horse, the two fields, the pathway,

The Interpretation of Dreams 149

and the wooden steps. The various elements in the dream
consequently might simply be memories of things seen but
not leaving any conscious impression. Nor can the dis-
similarities of the two stories be excluded, such as that in the
one case Dunne was alone and in the other that he was ac-
companied by his brother ; or that in the one case the horse
was running after Dunne down the pathway, whereas in the
actual event the horse was running by himself.

These are probably the most impressive of Dunne's
dreams, and there would be no point in quoting any others.
He was sufficiently encouraged by this evidence to perform a
series of experiments in which he wrote down in detail all
his dreams and then looked for events on the following days
which could be matched with these dreams. He also got
other people to do the same, and he claims that a large num-
ber of predictive dreams could be observed in this fashion.
When looked at dispassionately and with the knowledge
that those quoted in his book are presumably the most con-
vincing among the many thousands which Dunne could
find, the sceptical reader will see no good reason to renounce
his scepticism.

As a method of proof, Dunne's attempt breaks down be-
cause there is no way of estimating the influence of chance
factors, the admixture of previous acquaintance, and other
similar complicating features. He himself agrees that what
he has to tell is not proof in the ordinary sense and recom-
mends the reader to try the method for himself. This, he
claims, will lead to complete conviction. Even if this claim
were true, and I have recorded dreams for several months
without finding one which was in the slightest degree pre-
dictive of anything whatsoever, it would of course prove
nothing at all. Emotional certainty that a particular belief is
true still does not constitute proof, though it may be shared
by a large number of people. Dunne takes this notion of
proof altogether too lightly to make his work worthy of more
serious consideration than that of his many predecessors.

Only one well-controlled study has been reported in the
literature. This investigation was carried out in March 1932.

150 Part One • Borderlands of Knowledge

A few days after the kidnapping of the Lindbergh baby, but
before the body had been found, the investigators put in the
papers throughout the country a request for dreams con-
cerning the kidnapping, and obtained over 1,300 replies.
The dreams were then compared with the facts as es-
tablished several weeks later - namely, that the baby's naked
and mutilated body was discovered in a shallow grave in some
woods near a road, and that death had been instantaneous.
Only in some 5 per cent of the dreams sent in did the baby
appear to be dead, and in only seven dreams were the actual
location of the body, its nakedness, or the manner of its
burial reasonably accurately portrayed. Only four of the
seven dreams included the three items: death, burial in
a grave, and location among trees. The following dream was
judged to be the most accurate of the seven :

I thought I was standing or walking in a very muddy place
among trees. One spot looked as though it might be a round shal-
low grave. Just then I heard a voice saying 'The baby has been
murdered and buried there'. I was so frightened that I immedi-
ately awoke.

* When it is remembered that this is the most accurate
dream out of 1,300 and that it contained only a few of the
facts later established, we will hardly be impressed by the
precognitive abilities of the people who sent in their dreams.
It could of course be argued that only a few people have this
ability and that these people did not send in reports, but this
would still leave us with the task of discovering those people
who could supply us with the evidence. In summary, while
the notion of precognitive ability is an intriguing one, it does
not seem to find very much support from the study of dreams.
What is indeed more needed than anything is a method, such
as that discovered by Rhine, Soal, and others, in the case of
clairvoyance and telepathy, which will enable us to obtain a
reliable estimate of the chance factors entering into our ex-
periment. In the absence of such an estimate, it can always
be argued that coincidence is the basis of the reported find-
ings, and that chance coincidences should not be regarded

The Interpretation of Dreams 151

as evidence for something of such very high a priori unlikeli-
hood as dreaming the future even before it has happened. It
would be unscientific to deny in principle the possibility that
such evidence may ultimately be found, but it must be
concluded that up to now the evidence in favour of such a
view does not even begin to reach a reasonable level.

We must turn to the second and quite different type of
dream interpretation which is current, in order to see
whether this has any more evidence in its favour. As much
of this is tied up with the theory of Freud, a brief discussion
of this at least is essential, although most people will already
be familiar with some of its aspects. However, as in so many
cases, what Freud is popularly believed to have said is not al-
ways the same as what he actually did say, and it may serve a
useful function to set out the theory briefly and succinctly.

According to the Freudian theory and that of many others
whose writings have preceded his by hundreds or even thou-
sands of years, dreams do not reveal anything about the
future. Instead, they tell us something about our present un-
resolved and unconscious complexes and may lead us back
to the early years of our lives, when, according to psycho-
analytic theory, the ground was being prepared for these
later defects. There are three main hypotheses in this general
theory which will be discussed in some detail. The first hypo-
thesis is that the dream is not a meaningless jumble of
images and ideas, accidentally thrown together, but rather
that the dream as a whole, and every element in it, are
meaningful. The second point that Freud makes is that dreams
are always in some sense a wish fulfilment; in other words,
they have a purpose, and this purpose is the satisfaction of
some desire or drive, usually of an unconscious character.

Thirdly, Freud believes that these desires and wishes, hav-
ing been repressed from consciousness because they are un-
acceptable to the socialized mind of the dreamer, are not
allowed to emerge even into the dream without disguise. A
censor or super-ego watches over them and ensures that they
can only emerge into the dream in a disguise so heavy that
they are unrecognizable.

152 Part One • Borderlands of Knowledge

Let us look at these three propositions in turn. The idea
that the dream is meaningful is, as we have seen, a very
ancient one. For Freud it follows directly from the deter-
ministic standpoint: i.e. from the view that all mental and
physical events have causes and could be predicted if these
causes were fully known. This is a philosophical notion with
which few scientists would wish to quarrel. It should be
noted, however, that it is possible to believe that all natural
events, including dreams, are caused by some agency or
other without necessarily believing that dreams are mean-
ingful in the Freudian sense. The great neuro-surgeon Pen-
field has shown that when needles are pushed into certain
parts of the brain and an electric current sent through the
needle, whole memory sequences of a specified kind run
through the mind of the patient. The same memory se-
quence can be evoked again and again by simply re-apply-
ing the current. It might be possible that random stimula-
tion of a physical nature of different parts of the cortex pro-
duces random combinations of past memories which are
overlaid during the waking state by our sensory impressions,
but which during sleep come to the fore in the form of
dreams. This is not presented as a serious theory; it is merely
intended to illustrate that dreams could be deterministically
explained without having to accept their being in any sense
meaningful. This point has often caused confusion because
scientists, being generally in favour of determinism, have
accepted the Freudian explanation of the meaningfulness of
dreams by virtue of it being a deterministic explanation.
When it is realized that many other deterministic explana-
tions may be possible, we may be more willing to look at the
evidence in favour of this particular interpretation to see
whether in fact it is borne out.

Freud's argument of the meaningfulness of dreams is
directly connected with his general theory that all our acts
are meaningfully determined ; a theory which embraces mis-
pronunciations, gestures, lapses, emotions, and so forth. An
example may be found in the story told by Jean Jacques
Rousseau. Apparently he had acquired the habit of making

The Interpretation of Dreams 153

a detour as he approached a certain boulevard. On ques-
tioning himself with regard to the origin of this 'mechanical
habit', he came to the following conclusion: 'This is what
my reflection discovered, for none of it had until then been
present in my mind ' ; it was a matter of avoiding a little
beggar whose chatter he disliked. ' We have no mechanical
impulse,' Rousseau continues, 'the cause of which may not
be found in our hearts if we but knew how to seek it there.'
This is indeed the essence of the Freudian doctrine of mean-
ing, a doctrine which was widely accepted and discussed
hundreds of years before Freud was born, but of which
he was an extremely effective popularizer.

Let us now turn to the second part of Freud's doctrine,
namely the view that the dream is always a wish fulfilment. -
This is linked up with his general theory of personality.
Roughly speaking, Freud recognized three main parts of
personality : one, which he calls the id, is a kind of reservoir
of unconscious drives and impulses, largely of a sexual
nature; this reservoir, as it were, provides the dynamic
energy for most of our activities. Opposed to it we have the
so-called super-ego, which is partly conscious and partly un-
conscious and which is the repository of social morality.
Intervening between the two, and trying to resolve their
opposition, is the ego: i.e. the conscious part of our personal-
ity. In religious language we might liken the id to the con-
cept of original sin and that of the super-ego to the concept of
conscience. Classical scholars will have no need to be re-
minded of the striking anticipation of the Freudian doctrine
which is contained in Plato's Phaedrus. This is what Socrates
tells us in his story:

'As I said at the beginning of this tale, I divided each soul into
three - two horses and a charioteer ; and one of the horses was good
and the other bad ; the division may remain, but I have not yet ex-
plained in what the goodness or badness of either consists, and to
that I will now proceed. The right-hand horse is upright and
cleanly made; he has a lofty neck and an aquiline nose; his colour
is white and his eyes dark ; he is a lover of honour and modesty and
temperance, and the associate of right opinion ; he needs no touch

154 Part One ' Borderlands of Knowledge

of the whip, but is guided by word and admonition only. The
other is a crooked lumbering animal, put together anyhow ; he has
a short thick neck; he is flat-faced and of a dark colour, with grey
and blood-shot eyes ; the mate of insolence and pride, shag-eared
and deaf, hardly yielding to whip and spur. Now when the
charioteer beholds the vision of love, and has his whole soul
warmed through sense, and is full of the prickings and ticklings of
desire, the obedient steed, then as always under the government of
shame, refrains from leaping on the beloved; but the other, heed-
less of the pricks and of the blows of the whip, plunges and runs
away, giving all manner of trouble to his companion and the
charioteer, whom he forces to approach the beloved and to remem-
ber the joys of love.'

As one commentator reflects, ' One could hardly hope to
find a more beautiful description of the three main com-
ponents of man's personality - his primitive, vital impulses,
his conscious or idealized self and his reason'. This part of
Freud's theory too is then by no means novel and original,
but has been part of educated thought for over 2,000 years.

The link-up between Freud's theory of personality and his
theory of dream interpretation is a very simple one: the
forces of the id are constantly trying to gain control of the
ego and to force themselves into consciousness. During the
individual's waking life, the super-ego firmly represses them
and keeps them unconscious; during sleep, however, the
super-ego is less watchful, and consequently some of the de-
sires start up in the id and are allowed to escape in the form

- of dreams. However, the super-ego may nod, but it is not
quite asleep, and consequently these wish-fulfilling thoughts
require to be heavily disguised. This disguise is stage-
managed by what Freud calls the dreamwork. Accordingly,
it is necessary to distinguish between the manifest dream, i.e.
the dream as experienced and perhaps written down, and
the latent dream, i.e. the thoughts, wishes, and desires ex-
pressed in the dream with their disguises removed. The task
of the analyst and interpreter on this view is to explain the
manifest dream in terms of the latent dream.

— How is this to be done? Freud uses two methods. The first

The Interpretation of Dreams 155

of these we will welcome as an old friend ; it is simply the
method of symbolic interpretation which we have already
encountered in the case of Artemidorus. The other method, -
of much greater general interest and importance, is the
method of association.

First let us consider the use Freud makes of the theory of
symbolism. Very much like the old dream books, Freud pro-
vides whole lists of symbols standing for certain things and
certain actions. However, where the old dream books had
rather a catholic interest and spread their net widely, Freud —
concentrates almost exclusively on sex and sexual relations.
The male sex organ is represented in the dream by a be-
wildering variety of symbols. Anything that is long and
pointed - a stick, a cigar, a chimney, a steeple, the stem of a
flower - is so interpreted because of the obvious physical re-
semblance. A pistol, a knife, forceps, a gun - these may stand
for the penis because they eject and penetrate; similarly a
plough may become a sex symbol because it penetrates the
earth. Riding a horse, climbing stairs, and many, many
other common-sense activities stand for intercourse. Hollow
objects and containers are feminine symbols : houses, boxes,
saucepans, vases - all these represent the vagina.

Members of the family are frequently said to be sym —
bolically represented in the dream; thus the father and
mother may in the dream appear as king and queen.
There would be little point in giving long lists of objects and
acts and their symbolic interpretation in Freudian terms (we
shall encounter some of these in the course of our discussion
of some dreams later on). We may also postpone a critical
discussion of this Freudian theory and the elaboration of a
more reasonable alternative theory until we have looked at
the method of free association.

The technique of free association is essentially based on — -
the nineteenth-century doctrines of the associationist philo-
sophers. They believed that ideas became linked through
similarity or through contiguity and that mental life could
be understood entirely in terms of such associations. If ideas
are linked in a causal manner, as is suggested by this theory,

156 Part One • Borderlands of Knowledge

then we should be able to find links between manifest and
latent phenomena by starting out with the former and,
through a chain of associations, penetrate to the latter. In
other words, what is suggested is this : starting out with cer-
tain unacceptable ideas which seek expression, we emerge
finally with unintelligible ideas contained in the manifest
dream. These, having been produced by the original latent
ideas, are linked to them by a chain of associations, and we
shall be able to re-discover the original ideas by going back
over this chain of ideas. In order to do this, Freud starts out
by taking a single idea from the manifest dream and asking
the subject to fix that idea in his mind and say aloud any-
thing that comes into his mind associated with that original
idea. The hope is that in due course a chain of associations
will lead to the latent causal idea. An example of the process
of reassociation taken from M. F. Frink may illustrate the
method. The original point of departure in this case is not a
dream but a lapse of memory; however, from Freud's point
of view both types of mental facts are determined on similar
lines and are equally useful for illustrating the method.

One of Frink's friends asked him the name of a shop which
sold a certain article. Frink remembered the shop but found
that he had forgotten its name. Some days later, on passing
the shop, which he knew very well, he saw that the name
which he had not been able to remember was Pond. This is a
type of event which, according to Freud, is motivated and
has a meaning, and Frink decided to discover this meaning
by means of free association. Starting out with the word
Pond, he thought of a certain Doctor Pond who was a pitcher
in a baseball team. From there his thoughts went on to the
Indian pond -where, he used to go fishing as a child, and he saw
himself throwing into the water a big stone which served as
an anchor to his boat. Then he thought of a man called
Fisher, who also played as pitcher in a baseball team. Con-
tinuing his associations, he thought of Pond's Extract; this
product contains witch-hazel, and this reminded him that in
his childhood he used to rub his arm with witch-hazel when
he was pitcher in a baseball team. This reminded him of a

The Interpretation of Dreams 157

rather fat boy, a member of the same team, who had once
fallen head foremost into a muddy puddle and had come
out in so begrimed a state that he looked just like a pig. This
brought up the memory of another young man nicknamed
Piggy, and Frink then recalled that he himself had been
nicknamed pig.

His associations were interrupted for a few moments and
he started again with the word pond. This produced the
word ponder; then think; then the expression 'sicklied o'er with
the pale cast of thought'. This led him to Hamlet; to a certain
village which he had thought of as a hamlet, and to a farmer
in that village who had told him that one of his neighbours
had in sheer malice killed two pigs and had thrown them
into the farmer's well. At that point a memory of his seventh
year arose suddenly which gave Frink the explanation of his
failure to remember. He used to play with his brother on the
edge of a pond, and a dog, of which he was very fond, used
to swim in the pond. The boys used to throw stones to the
dog, which the animal tried to catch. At one time the young
Frink, wishing to frighten the dog, threw her a large stone,
taking bad aim. The stone struck the dog squarely on the
nose and she drowned. This was a terrible grief to the little
boy and for months he was quite inconsolable. The memory
of the accident sometimes made him cry out during the
night. The psychoanalytic explanation, then, is that the
word pond was not forgotten by chance, but that it had been
thrust out of consciousness because of its painful connexion
with the drowning of the dog.

Our main interest here centres on the associations pro-
duced by Frink. From the beginning these seem to be
grouped around a focal point which is not seen at the outset,
but which suddenly appears at the end. Take thefirst associa-
tion - that relating to Doctor Pond. Frink himself is a doctor,
the accident took place beside a pond, and the pitcher in base-
ball is a man who throws a ball. The next association is that
of the Indian pond; this pond is situated in the same town as
that in which the dog drowned, and Frink appears in it as
throwing a big stone. The third association introduces a

158 Part One • Borderlands of Knowledge

certain Fisher (fisherman) and again the idea of throwing. The
associations leading off from the reference to Hamlet still con-
tain the idea of throwing and of falling into the water, but
this is not all. There is the appearance of the young man
nicknamed Piggy and the killing of the two pigs. This may
appear obscure until it is realized that the dog's name was
Gip, which is the word pig in reverse.

This is a typical example of interpretation by means of
free association, and the reader will probably agree that the
method has a certain interest. The story tries to prove the
existence of a causal relationship between the forgetting of
the word Pond and Frink's childhood experience, but this, of
course, it does not succeed in doing. The only thing it does
prove, namely that terms similar in meaning tend to be asso-
ciated in one's mind, hardly requires so intricate and elabor-
ate a proof. Nevertheless, the idea of using the method of
association in exploring the contents of the mind is a highly
original and brilliant one, and much credit must go to the
man who first introduced it into psychology. This man, con-
trary to popular belief, was not Freud, however, but Sir
Francis Galton. Galton's name is probably best known
through his initiation of the Eugenics movement, but his
claims to fame extend in many different directions. He was
one of the most versatile and brilliant of the nineteenth-
century scientists, and his discoveries in many fields have
considerably influenced our lives. He has many claims to be
called the founder of modern psychology, a title usually be-
stowed on a plodding, methodical German called Wundt,
whose contributions were of an administrative rather than
of a creative nature.

Galton was one of the last universal geniuses, and he in-
vestigated and made major contributions to a great variety
of subjects. One of his biographers gives the following list :
* Travel, the weather, stereoscopic maps, high-pitched
whistles, blood transfusions, composite photography, finger
prints, number forms and word association, correlation,
twins, the sterility of heiresses, and various contrivances and
inventions'. The importance of some of this work may be

The Interpretation of Dreams 159

gathered from what the great scientist and statistician Pear-
son has to say about Galton's contribution to one of these
fields, namely that of correlation. Talking about some em-
pirical investigation of Galton's, Pearson has this to say :

The need for novel statistical methods which its problems de-
manded, led him to the correlational calculus, the fons et origo of
that far reaching ramification, the modern mathematical theory of
statistics. . . . From that conception arose a new view of the universe
both organic and inorganic which provides all science with a
novum organum far wider reaching in its effects than that of Bacon,
and it is as characteristic of the last quarter of the nineteenth cen-
tury as the fluxional calculus was of that of the seventeenth.

Galton tried out on himself an elaborate system of word
association tests and reached conclusions very similar to
those later popularized by Freud and Jung. The results, he
said,

gave me an interesting and unexpected view of the number of the
operations of the mind and of the obscure depth in which they took
place, of which I had been little conscious before. The general im-
pression they have left upon me is that which many of us have ex-
perienced when the basement of our house happens to be under
thorough sanitary repairs and we realize for the first time the com-
plex system of drains and gas and water pipes, flues, bell wires, and
so forth, upon which our comfort depends, but which are usually
hidden out of sight, and with whose existence, as long as they acted
well, we had never troubled ourselves.

Making use, then, of these methods of symbolic inter-
pretations and of association, both discovered long before
his time, Freud proceeded to analyse the nature of the
dream. He discusses his discoveries in terms of so-called-
mechanisms which are active in the dream. The first of these
mechanisms he calls that of dramatization. This simply de-
notes the fact, already familiar to most people, that the
major part in dreams is played by visual images, and that
conceptual thought appears to be resolved into some form of
plastic representation. Freud likens this to the pictorial man-
ner in which cartoons portray conceptual problems. The

160 Part One • Borderlands of Knowledge

cartoonist is faced with the same difficulty as the dreamer. He
cannot express concepts in words, but has to give them some
form of dramatic and pictorial representation. He cannot
say in so many words that Kaiser Wilhelm the Second of
Germany was unwise in dismissing the experienced Chan-
cellor Bismarck and in taking the conduct of affairs into his
own hands, but he can dramatize this in pictorial terms
along the lines of the famous Punch cartoon entitled ' Drop-
ping the Pilot', in which the maniacal Emperor is seen on
the bridge of the ship, while the sad-faced pilot (Bismarck)
is seen descending the gangway.

- In addition to visual images, verbal ones also may appear,
and here the material meaning of words may often be asso-
ciated with a rather uncommon meaning. Long before
Freud, the Marquis Hervey de Saint-Denis, one of the most
acute students of the dream, had observed that the use of
verbal images by the dream-thought may sometimes have
the appearance of the pun. As an example he quotes an
occasion where the word 'Rosalie' aroused in the dreamer
the image of a bedspread, upon which roses were em-
broidered. As a pun, this connexion between the name
'Rosalie' and the words Rose-a-lit (rose in bed) is not a good
one, but it is quite typical of the way the dramatization pro-
cess makes use of verbal images.

Closely related to the mechanism of dramatization is that
of symbolism, which we have already encountered. Here is
an example which will illustrate the general mechanism of
dramatization and also that of symbolism in dreams. In this
dream a young woman dreamed that a man was trying to
mount a very frisky small brown horse. He made three un-
successful attempts; at the fourth he managed to take his
seat in the saddle and rode off. Horse-riding, as was men-
tioned earlier, often represents coitus in Freud's general
theory of symbolism. What happens when we look at the
subject's associations? The horse reminded the dreamer that
in her childhood she had been given the French word
' chevaV as a nickname; in addition, this woman was a small
and very lively brunette, like the horse in the dream. The

The Interpretation of Dreams 161

man who was trying to mount the horse was one of the
dreamer's most intimate friends. In flirting with him she had
gone to such lengths that three times he had wished to take
advantage of her but each time her moral sentiments had
got the upper hand at the last moment. Inhibitions are not
so strong in the dream, and the fourth attempt therefore
ended in a wish fulfilment.

Another mechanism acting in the dreamwork is said to be
that of condensation. The manifest content is only an abbre-
viation of the latent content. As Freud puts it: 'The dream is
meagre, paltry, and laconic in comparison with the range
and copiousness of the dream thoughts.' The images of the
manifest content are said by Freud to be over-determined:
i.e. each manifest element depends on several latent causes
and consequently expresses several hidden thoughts. As an
example, here is a dream reported by a young woman. In
this dream she was walking on Fifth Avenue with a friend,
and stopped in front of a milliner's shop window to look at
hats; at last she went in and bought one. Frink, who ana-
lysed this dream, obtained the following series of associa-
tions: The woman had actually been walking on Fifth
Avenue with her friend, but had not bought any hats. Her
husband had been in bed that day, and although his illness
was not serious, she had been very uneasy and could not get
rid of the notion that he might die. During the walk the
young woman had talked of a man whom she had known
before her marriage and with whom she had been in love.
Asked why she had not married him, she explained that his
financial and social position had been too far above her own.
When asked to associate on the buying of the hat, she men-
tioned that she had much admired the hats in the milliner's
shop window and would very much have liked to buy one,
but could not do so because of her husband's poverty.

Clearly, says Frink, the dream was satisfying her wish by
allowing her to buy a hat. However, that is not all; she sud-
denly remembered that in her dream the hat which she
bought had been a black hat: a mourning hat. Frink's inter-
pretation of this fact is as follows. The day before the dream,

1 62 Part One • Borderlands of Knowledge

the woman was afraid her husband would die ; in dreaming
that she was buying a mourning hat she was fulfilling a
death fantasy. In real life she was prevented from buying a
hat by her husband's poverty; in the dream she was able to
buy one. The implication is that she now has a rich husband,
and we see from her associations that indeed there is a very
rich man in the offing. Frink therefore concludes that the
young woman is tired of her husband, that her fear of seeing
her husband die is a defence reaction against her real wish
for his death, and that she would like to marry the man with
whom she was in love and who has enough money to satisfy
all her wishes. When Frink acquainted his patient with his
interpretation of her dream, she admitted that it was justi-
fied and told him several facts which confirmed it, the most
important of these being that after her marriage she had
learnt that the man with whom she had been in love had
also been in love with her; this had revived her feelings, and
she had regretted her hasty marriage. If we accept this inter-
pretation, we see how the buying of the hat is over-deter-
mined by the threefold wish : to see the death of her husband,
to marry the man she loved, and to have money.

The last dream mechanism which we will discuss is that
of displacement. It is a process whereby the emotional con-
tent is detached from its proper object and attached instead
to an unimportant or subsidiary one; the essential feature of
the latent content of the dream accordingly may sometimes
be hardly represented at all in the manifest content, at least
to outward appearances; it has been displaced under some
apparently innocuous object. Here is an example of such a
displacement in which a girl suffering from an obsessional
neurosis dreamed that she was in the presence of someone
whose identity was very vague but to whom she was under
some kind of obligation. As a token of her gratitude, she
made him a present of her comb. The relevant details of the
girl's history are as follows. She was a Jewess whose hand
had been sought in marriage a year earlier by a Protestant,
whose feelings she returned fully. However, the difference of
religion had prevented the engagement taking place, as the

The Interpretation of Dreams 163

girl believed that the arrival of children in such a mixed mar-
riage would cause discord and unhappiness because of the
problem of the religion in which they should be brought up;
accordingly she had refused her suitor. The night before the
dream, the girl had had a violent quarrel with her mother
and had thought it better for all concerned if she were to
leave home. She had gone to sleep thinking of ways and
means whereby she could support herself without being de-
pendent on her family. When asked to associate on the word
'comb', she answered that sometimes when someone was
about to use a brush or comb which did not belong to him,
she had heard it said, 'Don't do that; you will mix the
breed'. This association suggests the interpretation of the
dream; the person whose identity remains vague is the ex-
suitor; by offering him her comb, the girl shows her wish to
'mix the breed', i.e. to marry him and to bear his children.
The dream thus expresses an extremely important intention.
However, the mechanism of displacement has intervened so
that the idea of marrying the young Protestant is not recog-
nizable in the manifest content and is replaced by an appar-
ently irrelevant substitute, the gift of the comb.

We have now studied a few dreams as examples of the
Freudian method, and we must try to come to some form of
judgement on the adequacy of the Freudian hypothesis. It
should perhaps be explained that the dreams chosen as illus-
trations are very much clearer, more definite, and more in
line with Freudian theory than the great majority of dreams
found in the text-books of Freud himself and his followers.
They have all been quoted more than once by psycho-
analysts in support of their theories. In order to examine
Freud's hypothesis fairly, we have taken examples which
should support it as strongly as any ; if they do not bear out
his theory, the verdict must indeed be a negative one.

Now, it will be remembered that the central piece of
Freud's whole theory, the one bit that is original and not
derivative, is the notion that symbols and other dream
mechanisms are used to hide something so obnoxious, so
contrary to the morality of the patient, that he cannot bear

164 Part One • Borderlands of Knowledge

to consider it undisguised, even in his dream. This notion
seems so contrary to the most obvious facts that it is difficult
to see how it can ever have been seriously entertained. Let
us list some of these objections. In the first place, the notion
which is expressed symbolically in one dream may be quite
blatantly and directly expressed in another. We have a
highly symbolic and involved dream which is interpreted as
meaning that we want to kill a relative or have intercourse
with someone, only to find that in another dream these ideas
are expressed perfectly clearly in the sense that we do actu-
ally kill our relative or have intercourse with this girl. What
is the point of putting on the masquerade on one occasion,
only to discard it on another? As Jocasta said to Oedipus
'Many young men dream of sleeping with their mothers';
if they do, why should they at times go to the trouble of
dreaming that they are shooting off a revolver at a cow?

A second objection is that the symbols which are supposed
to hide the dream-thought very frequently do nothing of the
kind. Many people who have no knowledge of psycho-
analysis are able to interpret the sexual symbols which occur
in dreams without any difficulty at all. After all, let us face
the fact that there are many slang expressions in use refer-
ring to sexual activities and sexual anatomy, and that these
slang terms are only too often identical with Freudian sym-
bols. There seems to be little disguise in a person's dreaming
about a cock, symbolizing the penis, when the very same
person would not even know the term penis and always re-
fers to his sex organ as his 'cock'. Freud seems to have been
singularly remote from the realities of everyday life.

A last point of criticism has been raised by Calvin S.Hall,
whose theory we shall be considering in a minute. He asks
why there are so many symbols for the same referent. In his
search of the literature he found 102 different dream-
symbols for the penis, ninety-five for the vagina, and fifty-
five for sexual intercourse. Why, he asks, is it necessary to
hide these reprehensible referents behind such a vast array
of masks?

Let us see to what extent Freud's theory is in fact sup-

The Interpretation of Dreams 165

ported by the three dreams we have quoted. First of all let
us take the young lady nicknamed ' ChevaP, who at the last
moment frustrated three determined efforts at seduction by
her boy friend, only to have the success of the enterprise pre-
sented to her in a dream in symbolic form. According to the
Freudian theory, we would have to believe that the notion of
actually having intercourse with her boy friend was so
shocking to this young lady, and so much outraged her
moral instincts and training, that she could not even con-
template the idea in her sleep, thus having to disguise it in
symbols. This, surely, is a very unconvincing argument; to
imagine that a young girl, who would several times running
indulge in such heated love-making that she was on the
point of losing her virginity, could not bear to contemplate
the possibility of having intercourse, and had to repress it
into her unconscious, could surely not be seriously main-
tained, even by a psychoanalyst following obediently in the
steps of the master !

Much the same must be said with respect to the young
lady who dreamt that she bought a mourning hat. Here
again it is extremely unlikely that the ideas finding ex-
pression in symbolic form in the dream - I would like to
have money to buy clothes with; I would like to have mar-
ried the man I was in love with; I wish I could get rid of my
husband - were so abhorrent to the dreamer that she re-
fused to entertain them consciously or permit them open ex-
pression even in her sleep. In fact, the account given by the
analyst makes it clear that these ideas were quite con-
sciously in her mind ; why, then, did they require a symbolic
disguise?

The ' comb ' dream of the girl who did not want to ' mix
the breed' is an equally obvious contradiction to the Freud
hypothesis. The notion which finds disguised expression in
the dream - I would like to marry my fiance and get away
from my family - was quite acceptable to the dreamer, and
indeed was a constant source of conscious preoccupation in
her mind. There is no possibility here of arguing that the
latent dream-thoughts emerging from the id were so beastly,

1 66 Part One • Borderlands of Knowledge

sordid, and unacceptable that they had to put on a mas-
querade before being allowed to emerge even in a dream.

We thus find that the three dreams which have been
selected by psychoanalysts from many thousands as pro-
viding the most clear-cut support for psychoanalytic notions
are, in fact, a clear disproof of the major point of the Freud-
ian theory. We may, therefore, with some confidence, reject
Freud's hypothesis regarding the origin of the mummery
which takes place in dreams. It is unlikely enough on a
priori grounds, it is self-contradictory and it is not borne out
by psychoanalysis of specially selected dreams even.

Is it possible to substitute a more plausible theory for the
one that we have just rejected? An interesting move in this
direction has recently been made by C.S.Hall, a well-
known American psychologist. His argument is as follows.
The same objective fact - say sexual intercourse - may have
widely different meanings to different people. One concep-
tion might be that of a generative or reproductive activity;
another one might be that of an aggressive physical attack.
It is these different conceptions of one and the same objec-
tive fact which are expressed in the special choice of sym-
bolism of the dream. Dreaming of the ploughing of a field or
the planting of seeds is a symbolic representation of the sex
act as being generative or reproductive. Dreaming of shoot-
ing a person with a gun, stabbing someone with a dagger, or
running down with an automobile, symbolizes the view of
the sex act as an aggressive attack. According to this theory,
symbols in dreams are not used to hide the meaning of the
dream, but quite on the contrary, are used to reveal not only
the act of the person with whom the dreamer is concerned,
but also his conceptions of these actions or persons.

If a person dreams of his mother, and if his mother in the
dream is symbolized by a cow or a queen, Freud would inter-
pret this to mean that the dreamer is disguising his mother
in this fashion because he cannot bear to reveal, even to him-
self, the wishes and ideas expressed in the dream and con-
nected with the mother figure. In terms of Hall's theory, the
interpretation would be that the dreamer not only wishes to

The Interpretation of Dreams 167

represent his mother, but also wants to indicate that he re-
gards her as a nurturent kind of person (cow), or a regal and
remote kind of person (queen). The use of symbols, then, is
an expressive device, not a means of disguise, and it is note-
worthy that in waking life, symbols are used for precisely the
same reason: a lion stands for courage, a snake for evil,
and an owl for wisdom. Symbols such as these convey
in terse and concise language abstruse and complex con-
ceptions.

Certain symbols, on this theory, are chosen more fre-
quently than others because they represent in a single object
a variety of conceptions. The moon, for instance, is such a
condensed and over-determined symbol of woman; the
monthly phases of the moon resemble the menstrual cycle ;
the filling out of the moon from new to full symbolizes the
rounding out of the woman during pregnancy. The moon is
inferior to the sun; the moon is changeable like a fickle
woman, while the sun is constant. The moon controls the
ebb and flow of the tides, again linking it to the family
rhythm. The moon, shedding her weak light, embodies the
idea of feminine frailty. Hall concludes : ' Rhythm, change,
fruitfulness, weakness, submissiveness, all of the conven-
tional conceptions of woman, are compressed into a single
visible object.'

To indicate the way in which Hall uses this theory to aid
in the interpretation of dreams, we may quote an example
from his book:

I was at a gas station where I worked, and my friend Bob was
there also working. It seemed to me that he was new and inexperi-
enced at the job because I was watching him check the oil on a car.
He pulled out the oil dip-stick and looked at it. At this point I went
up to him, rather angrily, and said 'Bob, in order to check the oil
you have to wipe off the oil on the dip-stick first and then put the
dip-stick back in, pull it out, and then get a reading'. He thanked
me for my help and the dream ended.

When the dreamer was asked to say what came into his
mind in connexion with the dream, the dreamer said it

1 68 Part One • Borderlands of Knowledge

reminded him of sexual intercourse and that Bob was
not doing it properly. Apparently Bob went around with
prostitutes, which the dreamer felt was wrong, and he wished
that Bob would stop doing this. The action of inserting the
dip-stick into the oil hole was a direct representation of inter-
course, and revealed a pretty mechanical conception of
sexual gratification. It will be obvious again how inappro-
priate the Freudian theory would be to this dream.

Hall's conception of dream symbolism seems to apply
quite well to many of the dreams he quotes. It does not seem
to apply particularly well to many other dreams ; the reader
may judge for himself whether it applies to the three dreams
previously quoted. The truth appears to be that any writer on
dream interpretation seems to find it possible to quote a few
dreams in support of his views, but that these theories can-
not usually be applied to dreams quoted by people having a
different theoretical outlook. This suggests that all theories
of dream interpretation may have a certain limited amount
of truth in them, but that they do not possess universal signi-
ficance, and apply only to a relatively small part of the field.
This conclusion is strengthened when it is further realized
that quite probably the person whose dreams are being ana-
lysed begins to learn the hypothetical symbolic language of
the analyst and obediently makes use of it in his dreams.
This may account for the fact that Freudian analysts al-
ways report that their patients dream in Freudian symbols,
whereas analysts who follow the teaching of Jung report
that their patients always dream in Jungian symbols, which
are entirely different from the Freudian !

There is one further difficulty in accepting the symbolic
interpretations presented by so many dream interpreters.
How, it may be asked, do we know that a motor-car stands
for the sexual drive; might it not simply stand for a motor-
car? In other words, how can the poor dreamer ever dream
about anything whatsoever, such as a house, a screw, a
syringe, a railway engine, a gun, the moon, a horse, walking,
riding, climbing stairs, or indeed anything under the sun, if
these things are immediately taken to symbolize something

The Interpretation of Dreams 169

else? What would happen if you took a very commonplace,
everyday event such as a train journey and regarded it as an
account of a dream? The reader will see in the following
paragraph how such a very simple and straightforward de-
scription is absolutely riddled with Freudian symbols of one
kind or another. Relevant words and phrases have been
italicized to make identification easier.

To begin with, we pack our trunks, carry them downstairs, and
call a taxi. We put our trunks inside, then enter ourselves. The
taxi surges forward, but the traffic soon brings us to a halt and
the driver rhythmically moves his hand up and down to indicate
that he is stopping. Finally we drive into the station. There is
still time left and we decide to write a postcard. We sharpen a
pencil, but the point drops off and we test our fountain pen by
splashing some drops of ink on to the blotting-paper. We push the
postcard through the slot of the pillar-box and then pass the bar-
rier and enter the train. The powerful engine blows off steam and
finally starts. Very soon, however, the train enters a dark tun-
nel. The rhythmic sounds of the wheels going over the inter-
sections send us to sleep, but we rouse ourselves and go to the
dining-car, where the waiter pours coffee into a cup from a long-
nosed coffee-pot. The train is going very fast now and we bob up
and down in our seats. The semaphore arms on the signal masts rise
as we approach and fall again as we pass. We look out of the
window and see cows in the pasture, horses chasing each other,
and farmers ploughing the ground and sowing seeds. The sun is
setting now and the moon is rising. Finally the train pulls into
the station and we have arrived.

It will be clear to the reader that there is practically noth-
ing that we can do or say on our journey which is not a
flagrant sex symbol. If, therefore, we wanted to dream of a
railway journey, the thing would just be impossible. All we
can ever dream about, if we follow the Freudian theory, is
sex, sex, and sex again. The reader may try the experiment
of describing a football match or a walk in the country or a
day at the office, without the use of phrases which would,
according to Freud, have a sexual connotation. He would
soon find that there is practically no object in common use,

1 70 Part One • Borderlands of Knowledge

and no activity frequently indulged in, which cannot be
made to symbolize some aspect of the sexual process.

The critical reader may feel at this point that while the
discussion may have been quite interesting at times, it has
not produced a single fact which could be regarded as hav-
ing scientific validity. Everything is surmise, conjecture, and
interpretation; judgements are made in terms of what seems
reasonable and fitting. This is not the method of science.
You do not argue about Ohm's Law or the Law of Gravita-
tion or the circulation of the blood. You state a definite
hypothesis, make certain deductions from this hypothesis,
and then proceed to carry out experiments to prove or dis-
prove your theory. That is the scientific method, and that is
precisely what is missing in all the work we have been sum-
marizing so far.

The blame for this state of affairs must be squarely laid at
the door of the analysts, whose efforts have always been
directed towards persuasion and propaganda, rather than
towards impartial investigation and proof. The reader may
recall our discussion of the necessity of having control groups
in psychological investigations. No control group has ever
been used in experimental studies of dream interpretation
by psychoanalysts, yet the necessity for such a control would
be obvious on reflection. According to Freud's theory, the
manifest dream leads back to the latent dreams in terms of
symbolization and in terms of free association. This is used
as an argument in favour of the view that the alleged latent
dream has caused the manifest dream, but the control ex-
periment is missing. What would happen if we took a dream
reported by person A and got person B to associate to the
various elements of that dream? Having performed this ex-
periment a number of times, I have come to the conclusion
that the associations very soon lead us back to precisely the
same complexes which we would have reached if we had
started out with one of person B's own dreams. In other
words, the starting point is quite irrelevant; as all roads lead
to Rome, so a person's thoughts and associations tend to
lead towards his personal troubles, desires, and wishes of the

The Interpretation of Dreams 171

present moment. I am not suggesting that this is an estab-
lished fact; I am merely suggesting that here is an alterna-
tive theory which will equally easily account for the observed
facts, but which has never been tested by psychoanalysts.
Many other alternative theories could be formulated and
would have to be tested before anything decisive could be
said about the value of the Freudian hypothesis. In the ab-
sence of such work, our verdict must be that, interesting as
some of these speculations are, such evidence as there is
leads one to agree with the many judges who have said that
what is new in the Freudian theory is not true, and what is
true in it is not new. It is possible that future experiments
may lend greater credence to psychoanalytic postulates in
this field, but for the moment we can only judge on the
evidence so far presented.

Actually it would not be quite correct to say that no ex-
perimental work on dreams had been done. There are a
number of promising leads, but, as might have been ex-
pected, these have come from the ranks of academic psycho-
logists and not from psychoanalysts themselves. Of particu-
lar interest is the work of Luria, a Russian psychologist who
attacked the problem of dream interpretation as part of a
wider problem, namely the experimental investigation of
complexes. His technique consisted in implanting complexes
under hypnosis and observing the various reactions, includ-
ing dreams, of the subjects after they had recovered from the
hypnotic trance. The implanted complexes were of course
unconscious in the sense that the subject knew nothing about
them on being interrogated, and had no notion of anything
that had transpired during the hypnotic trance. An example
may make clearer just what the procedure is. It is taken from
a study of my own, carried out to check some of the findings
reported by Luria. The subject, a thirty-two-year-old
woman, is hypnotized and the following situation is power-
fully impressed upon her as having actually happened. She
is walking across Hampstead Heath late at night when sud-
denly she hears footsteps behind her; she turns and sees a
man running after her; she tries to escape but is caught,

172 Part One • Borderlands of Knowledge

flung to the ground, and raped. On waking from the hypno-
sis, she is rather perturbed, trembles a little, but cannot ex-
plain the cause of her uneasiness ; she has completely for-
gotten the event suggested to her under hypnosis. She is then
asked to lie down and rest; after a few minutes she falls into
a natural sleep, but is immediately woken up and asked to
recall anything she might have been dreaming of. She re-
counts that in her dream she was in some desolate spot
which she cannot locate and that suddenly a big Negro,
brandishing a knife, was attacking her; he managed to prick
her thigh with it. The symbolic re-interpretation of the
hypnotic trance in the dream is clear enough and tends to
substantiate the fact that dreams express in dramatized and
symbolic form certain thoughts which in the waking state
would probably be conceptualized in a more direct form.
This method of investigation has considerable promise, but
unfortunately very little has been done with it. One well-
designed experiment would be worth more than all the
thousands of anecdotal contributions published in the
psychoanalytic journals, but unfortunately the trend has
been towards more and more uncontrolled, unconvincing,
and unrepeatable single-case studies, rather than towards a
properly controlled scientific and experimental kind of in-
vestigation.

Realizing, then, that nothing certain is known, can we at
least propound a general theory which summarizes what we
have said and is not contradicted by any of the known facts?
Such a theory might run as follows : The mind tends to be
constantly active. In the waking state most of the material
for this activity is provided by perceptions of events in the
outer world; only occasionally, as in problem-solving and
day-dreaming, are there long stretches of internal activity
withdrawn from external stimulation. During sleep such ex-
ternal stimulation is more or less completely absent, and
consequently mental activity ceases to be governed by ex-
ternal stimulation and becomes purely internal.

In general this mental activity is very much concerned
with the same problems that occupy waking thought. Our

The Interpretation of Dreams 173

wishes, hopes, fears, our problems and their solutions, our
relationships with other people - these are the things we
think about in our waking life, and these are the things we
dream about when we are asleep. The main difference is
that mental activity in sleep appears to be at a lower level of
complexity and to find expression in a more archaic mode
of presentation. The generalizing and conceptualizing parts
of the mind seem to be dormant, and their function is taken
over by a more primitive method of pictorial representation.
It is this primitivization of the thought processes which leads to
the emergence of symbolism, which thus serves very much
the function Hall has given it in his theory.

This symbolizing activity is, of course, determined to a
large extent by previous learning. To an Eskimo it would be
impossible to dream of ploughing as a symbol of intercourse
because Eskimos are not acquainted with the plough. To
the patient who is being analysed by a follower of Freud, it
would not occur to dream in Jungian symbols because he
has not become acquainted with them. In general, symbols
are relative to the education and experience of the dreamer,
although certain symbols, such as the moon, are very widely
used because they are familiar to almost all human
beings.

Would it be true, then, to say that dreams are useful in the
treatment of mental disorders? If we are correct in believing
that dreams are simply a 'continuation of thinking by other
means' (to paraphrase a famous saying), then they should
be able to tell us something about the problems, wishes, and
fears of the dreamer. However, it is probable that we would
get as much information, and more easily, by asking him
directly. The possibility remains that in a small number of
cases such revelation is impossible because the patient is truly
unconscious of his complexes and thoughts. If it could be
demonstrated that under such conditions dreams do, in fact,
reveal this unconscious material, then we would have to an-
swer our question in the affirmative. Until evidence of a
more rigorous kind than is available now is produced in
favour of this hypothesis, we can only say that no confident

174 Part One ' Borderlands of Knowledge

answer can be given. If and when the proper experiments
are performed, due care being given to the use of control
groups and other essential safeguards, then we shall be
able to end this chapter on a more positive note. At the
moment the only fit verdict seems to be the Scottish one of
'not proven'.

PART TWO

PERSONALITY AND SOCIAL LIFE

5

CAN PERSONALITY BE MEASURED?

This question, so often asked of the psychologist, is almost
completely meaningless. The answer depends on what we
mean by personality, what we mean by measurement, and,
indeed, one might even maintain that it depends on the
meaning of the term 'can'. Do we understand this to imply
that it is possible at the moment to carry out such measure-
ment, or do we mean that while it is impossible at the mo-
ment to do this, in principle it can and will be done? Before
going into any details, therefore, we must first of all have a
brief look at what is meant by 'personality' and what mean-
ing the term 'measurement' has in this connexion.

It is not an easy task to define personality. Allport, who
has written the classic introduction to this field, discusses
some fifty definitions without doing more than scratching
the surface. It would certainly not help us very much in our
quest to know that, to the philosopher Kant, 'personality
exhibits palpably before our bodily eyes the sublimity of our
nature'; that Stern defined personality as 'a multiform
dynamic unity'; or Windelband as 'individuality which
has become objective to itself. We may admire without
understanding, but on the whole we will probably agree
with a famous sociologist; according to him, 'This word
persona has rolled along with wonderful bounds, striking
right and left, suggesting new thoughts, stirring up clouds of
controversy, and occupying to the present day a prominent
place in all discussions on theology and philosophy, though
few only of those who use it know how it came to be there.'

i75

1 76 Part Two • Personality and Social Life

Fortunately, although we may have difficulty in defining
it, we can at least tell how the word 'personality' came to be
there. Persona originally referred to the theatrical mask used
first in Greek drama and later on adopted about the year
100 B.C. by Roman actors. It is said that this importation
was due to the fact that a popular Roman actor wanted to
hide an unfortunate squint behind these masks. In the writ-
ings of Cicero the word was then used to denote how a per-
son appears to others, but not as he really is, and also a
conglomeration of personal qualities. Very roughly, this
combination of meanings, referring both to the inner
psychological qualities of a person, and also to the impression
he gives to other people, has persisted and is contained in
our modern psychological usage of the term * personality'.

As such, however, it is far too all-embracing to be particu-
larly useful. It seems to be almost co-extensive with be-
haviour, and therefore with psychology as a whole, which is
usually defined as a study of behaviour. Can we nail it down
a little more precisely? We can do so, but only by ceasing to
regard it as a strict scientific term which can have a precise
meaning, such as atom, reflex, planet, acid, or molecule, and
treating it instead as a term descriptive of a field of study. In
this way we shall be able to get a much better notion of
what the exact place of personality is in modern psychology.

We may start out by, first of all, looking at the place of
psychology as a whole among the sciences. We will find that
it derives its unique importance from the fact that it acts as
a kind of bridge between two very large and important
groups of disciplines. On the one hand, it is closely related
to the biological sciences, like physiology, neurology,
anatomy, biochemistry, zoology, genetics, and so forth;
many of its theories are built on the findings of these sciences,
and any advance in them is immediately reflected in psycho-
logical research and theorizing.

On the other hand, psychology is intimately related to the
social studies, like sociology, economics, history, anthropo-
logy, social philosophy, psychiatry, and so on. These are
sometimes called the social sciences, but it seems to me that

i

Can Personality be Measured?

177

at their present stage of development the term science is a
little misleading as applied to fields characterized more by
speculation and blind empiricism than by experimentally
derived laws of general applicability. However that may be,
clearly these various fields of study depend very much on the
existence of a well-developed body of knowledge relating to
the behaviour of individuals. A properly developed science
of psychology is indispensable to their development, and the

Econometrics

Biological
Stienccs

.Social
Studies

Figure 3

need for such a science is clearly evidenced by the fact that
in its absence many of them have been forced to develop
their own ad hoc and frequently absurd systems of what one
can only call quasi psychology. The concepts of ' economic
man' and 'rational man' are two of the attempts made by
economists and sociologists to lay some kind of psychological
foundation for their studies.

The picture, in fact, is very much like that shown in
Figure 3, which clearly demonstrates the bridge-like position
of psychology. However, this notion of psychology as a uni-

1 78 Part Two • Personality and Social Life

fied subject standing between the biological sciences and the
social studies is not quite accurate. Psychology itself, as will
be seen in the diagram, is divided into two; on the one hand,
physiological psychology, which has its main affinities with
the biological sciences, and on the other hand, social psycho-
logy, which has its main affinities with the social studies. The
gap between these two groups of studies is a very deep one.
Physiological psychologists publish their papers in one set of
journals; social psychologists publish theirs in another.
There is little reference from one group of journals to the
other, and few readers are equally familiar with both sets.
Consequently, the gulf between the biological sciences and
the social studies runs right through the centre of psycho-
logy.

Within the discipline of psychology, then, clearly a con-
cept is needed to bring these two sides together. This central
role falls to the concept of personality, which thus acquires
its unique importance in psychology by bringing together
two sets of workers otherwise isolated from each other. It
does this, not so much by emphasizing a special field of
study, but rather by imparting a certain point of view to
groups of studies which otherwise might have remained
quite unrelated. An example may illustrate the difference
which a change in point of view of this kind may make.

Physiologists for many years have been interested in defec-
tive dark vision. They have shown that there are physio-
logical causes for it, some connected with the structure of the
retina; others with dietary deficiencies. During the First
World War several governments showed a considerable in-
terest in the diagnosis and the distribution of night blindness
because, obviously, soldiers suffering from this disability
could not be used for certain purposes and might, if not
diagnosed in time, become a danger both to themselves and to
their comrades. Consequently, physiologists and physiologi-
cal psychologists were asked to make estimates of the num-
ber of cases to be expected to fall into this category of the
night blind. In fact, these estimates turned out to be absurd
under-estimates of the true number of people who actually

Can Personality be Measured? 179

had visual defects in the dark, and the question arose as to
the cause of this very considerable discrepancy.

The answer became clear when it was found that many of
the people who were suffering from dark blindness also
seemed to be somewhat emotionally unstable individuals.
Detailed investigations disclosed that in many people who
were, in fact, night blind, there appeared to be no physio-
logical causes of any kind which might be responsible for
their condition. In these cases particularly, the emotional in-
stability and presence of neurotic symptoms were very
marked, and several experimental studies have since dis-
closed that there is quite a close relationship between the
difficulty of seeing in the dark and the possession of certain
neurotic personality traits. Thus, what caused the original
estimates to be so much in error was the fact that physio-
logists and their psychological colleagues treated the par-
ticular part of the central nervous system which subserves
dark vision in isolation. They dealt with the various parts of
the eye, the retina, the optic nerve, and so forth, in complete
disregard of the individual person of whom these various
structures formed a part. This, of course, is in the best tradi-
tion of physiology, which tries to isolate certain structures
and study these under conditions, as far as possible, removed
from any disturbances by other parts of the central nervous
system.

This type of segmental analysis is certainly valuable and
important, and nothing that is said here should be con-
strued as being in any way critical of work of this type.
However, it must be realized that in the intact animal, or in
the living human being, such interaction between different
structures is not the exception, but the rule. Consequently,
the study of segmental processes in complete isolation does
not tell us very much about the behaviour of the complete
organism. The physiological study has to be supplemented,
and it is here that the concept of personality comes in. At the
moment that physiological or any other processes are
studied from the point of view of their interaction with other
parts of the individual they become part of the study of

1 80 Part Two • Personality and Social Life

personality. Thus, personality study is not differentiated from
physiological psychology by the phenomena it studies, but
rather by the conditions under which these phenomena are
studied, and by the setting in which experiments are carried
out and information sought. Both the physiological psycho-
logist and the psychologist interested in personality study
may do work on dark vision. Where, however, the physio-
logical psychologist studies dark vision as far as possible un-
der conditions which exclude the influence of personality
factors, or what we may call 'central processes', to the per-
sonality oriented psychologist the phenomena of dark vision
are of interest precisely to the extent to which they are in-
fluenced by these 'central processes'.

The reason for this will be obvious. If we are interested in
such 'central processes' as liability to fear, or anxiety, we
soon come up against the difficulty that there is no direct
way of measuring them in the present state of our know-
ledge. We are reduced to finding certain measurable
phenomena which show a relationship with the central
processes in which our real interest lies. To discover, there-
fore, that night blindness, when found in people not suffer-
ing from physiological defects, is closely related to liability
to fear and anxiety is of considerable importance because it
enables us to use a measurement of defective dark vision as a
measure, albeit an indirect and perhaps a not very satis-
factory one, of the central processes which really engage our
interest.

Much the same might be said of conditioning. Most
people nowadays are familiar with Pavlov's famous experi-
ment, in which he showed that by pairing the sound of a bell
and the giving of food to a dog, the latter could be ' con-
ditioned' to react by salivation to the bell alone, i.e. when no
food was given. This mechanism of conditioning, which has
frequently been studied by physiological psychologists, is of
interest to workers in the field of personality because it can
be shown that individual differences in the speed with which
such conditioned reflexes are formed have very important
correlates in the behavioural field, and can thus be used as

Can Personality be Measured? 181

explanatory concepts to account for differences in ' personal-
ity'. A later chapter will develop this line of argument in
detail.

So much for the relationship between personality and
physiological psychology. What about its relationship with
social psychology? Here the connexion is perhaps more ob-
vious. Social psychology, strictly speaking, deals with the be-
haviour of people in groups. It attempts to formulate laws
which will enable us to predict what people will do under
certain conditions when two or more are co-operating or
acting in competition, or interacting with each other in some
other way. Such laws of interaction must inevitably intro-
duce in some way certain variables related to the individual
persons entering into the relationship. Thus, social psycho-
logy is depending very much for fundamental information
on the student of personality.

As an example, we might perhaps mention here the
formation of social attitudes and voting behaviour in an
election. These clearly are social phenomena, and certain
general laws can be discovered regarding them. Yet it
would be impossible to give anything like a full account of
the events the social psychologist is interested in without
paying considerable attention to personality factors in the
people concerned. Again an attempt will be made in a later
chapter to spell these very brief comments out in greater de-
tail, and to show how, through the intermediary concepts of
personality study, the laws of conditioning can be used to
account for many of the phenomena of attitude formation,
voting behaviour, and so forth.

This discussion of the place of personality in psychology
generally may have done something to indicate roughly the
field of study with which we shall be concerned. No attempt
will be made now to give a clear-cut definition because it is
doubtful if any such definition will be universally accept-
able, or would, indeed, add very much to the reader's under-
standing of the topic we shall be dealing with. The contents
of this chapter should give a rough idea, at least, of the kinds
of topic psychologists who are interested in personality do,

1 82 Part Two • Personality and Social Life

in fact, study, and the kind of methods they use in their
study.

If personality is a field of study characterized by a par-
ticular point of view, then obviously we cannot 'measure'
personality, just as little as we can measure the universe. All
we can do is to measure certain aspects of personality or of
the universe. Our question, therefore, becomes modified
immediately. Unfortunately, the problem of measurement
is a very complex and somewhat technical one, and it would
be out of place to discuss it here in any detail. A few remarks
must suffice to indicate the general line of the argument.
Measurement has been defined as the assignment of
numerical values to objects or events according to rules.
This is a very broad definition, but it is difficult to see how
it could easily be narrowed down without excluding pro-
cesses universally regarded as constituting measurement,
and without introducing artificial breaks where, in actual
fact, none seem to exist. If this definition be accepted, and
it has been widely accepted among scientists in general,
then we can see at once that many of those who argue
that measurement of personality variables is impossible
do, in fact, themselves use measurement in their daily
work.

Thus, the psychoanalyst who apparently argues against
the possibility of measurement being applicable to the com-
plexities of the human being does not hesitate to say that Mr
Buxtehude has an unresolved Oedipus complex, while Miss
Schwangerschaft has not, but by thus assigning a numerical
value of one to the gentleman, and a numerical value of
zero to the lady, he has, in fact, carried out an elementary
form of measurement. At least, he has attempted to carry out
an elementary form of measurement ; we would still have to
inquire into the reliability and validity of this measurement,
as well as into its meaningfulness. Quite likely we would find
that this measurement, so called, would have very little rela-
tion to actual fact. But that, of course, is a different prob-
lem; whether the attempt was successful or not, it has in
fact been made, and consequently our psychoanalyst cannot

Can Personality be Measured? 183

argue that measurement is impossible in this field unless he
wants to condemn his own practice at the same time.

Much the same is true of a great deal that is done by psy-
chiatrists, sociologists, and others who are temperamentally
opposed to measurement. What, in fact, they do is not to
substitute something else for measurement, but to use ele-
mentary, inappropriate, and illogical forms of measurement
where complex, appropriate, and relevant methods are
available. Like Monsieur Jourdain, in Moliere's famous
play, who discovered with considerable surprise that he had
been talking prose all his life, many people in these fields
may be surprised to find that they have been carrying out
measurement most of their lives.

It cannot, therefore, be reasonably maintained that mea-
surement is impossible or inappropriate to the variables
which the student of personality is interested in. What I
think most people who make statements of this kind are
really trying to say is that complex forms of statistical mani-
pulation are not appropriate to the measurements which
are, in fact, made, and that only the roughest and simplest
kind of treatment is acceptable. This, however, is not a really
tenable argument; in fact, quite the opposite appears to be
true. Where measurement is very accurate, complex mathe-
matics are seldom needed because the relationships and
functions involved are quite clear and obvious. When there
is a considerable amount of error in the measurement, how-
ever, then it becomes very necessary indeed to use complex
statistical methods to get estimates of the amount of error in-
volved, and to tease out the complex relationships existing
between data. This sorting-out process is a difficult one re-
quiring considerable technical competence, and certainly no
one not possessing this competence is likely to make a
worth-while contribution to this issue.

Having now agreed, in broad outline, on what we mean
by * measurement' and what we mean by 'personality', let
us have a look at the methods used by psychologists and
psychiatrists in their endeavours to pin down the very
elusive qualities associated with individual human beings.

184 Part Two • Personality and Social Life

First and foremost, we find there a method which has been
used since the beginning of time, although only in recent
years has it become formalized and subjected to experi-
mental study. This is the method of rating, i.e. the attribu-
tion of traits, abilities, attitudes, and so forth to other people
on the basis of our observation of their behaviour. All of us
at some time or other have decided that Mr Meetglad is
sociable, Mrs Puddlebasin talkative, Mr Greaseproof
humorous, and Miss Groansharp critical. When we make a
statement of this sort, what we are doing in fact is to attri-
bute a trait of a certain kind to these persons. We do this
without realizing that we are carrying out a very primitive
type of analysis, followed by a very primitive kind of mea-
surement. The analysis leads us to posit the existence of a
given trait; the measurement makes it possible for us to
assign each person to one of two categories, i.e. those having
or not having that particular trait. We may, in fact, go
further than that and have not two but several categories.
Thus, we may say that some people have a very good sense
of humour; others have a good sense of humour; some are
average ; some have a poor sense of humour ; and some a
very poor sense of humour indeed. Thus, we would assign
people to one of five categories and thereby get a rather more
refined type of measurement. But in essence this refinement
is irrelevant and we may, for the purpose of this discussion,
disregard it.

What about the validity of measurements of this kind?
Unfortunately, the evidence is fairly strongly opposed to our
paying very much attention to ratings made along these
lines, particularly when they are made by persons who have
not been trained specifically to be aware of the many diffi-
culties and complexities involved.

In the first place, when we rate a person as having a good
sense of humour, what precisely do we mean by this? We
might mean:

(1) A person who laughs readily, i.e. one who is a merry
sort of person, happy, good-natured, and so forth.

Can Personality be Measured? 185

On the other hand we might mean :

(2) A person who has a ready wit and tells amusing
stories and makes other people laugh.

Or again, we might mean :

(3) A person who laughs about the same sort of things
that we laugh about and who therefore agrees with us
in his tastes.

Again, we might mean:

(4) A person who is capable of laughing at himself, i.e. a
person who is not pompous, conceited, or blown up
with his own importance.

These are only some of the ways in which the concept
'having a good sense of humour' was defined by competent
and highly intelligent people who were asked to say what
they meant by certain terms. It is clear that simply to know
that person A thinks that person B has a good or a poor sense
of humour does not tell us very much unless we know pre-
cisely what person A means by having a 'good sense of
humour'. He may be able to tell us, of course, but even
when he does so, it will often be found that he is mixing up
two or three, or even all four of the different possibilities dis-
cussed above, so that he may rate one person as having a
good sense of humour for one reason and another person for
another reason. All this would not matter if the various
traits associated with these various definitions were closely
related, in that a person who laughed readily was also the
one with a tendency to tell witty stories, to laugh at himself,
or to agree with the rater in the things he found funny.
However, an experimental study has shown that in actual
fact there is no great tendency for these different traits to
hang together; indeed, a person who has a tendency to tell
witty stories frequently does not laugh at all readily at such
stories told by somebody else, and he often finds it very diffi-
cult to laugh at himself. Thus, what appears to be a very
simple rating procedure in fact turns out to be a highly

1 86 Part Two • Personality and Social Life

complex process which is very difficult, if not impossible,
to interpret in any meaningful sense.

Even if there is some agreement on just what the trait
being rated is supposed to mean, it may nevertheless be
impossible ito base any meaningful rating on ordinary
observation. Thus, in one study a large group of neurotic
patients was seen by two or more psychiatrists each. In view
of the importance which the trait of suggestibility is often
thought to have in producing neurotic symptoms, and in
view of the considerable amount of discussion which it has
received in the psychiatric literature, one might have
thought that there should be fairly general agreement be-
tween psychiatrists as to whether a given person was or was
not suggestible. Yet in actual fact it was shown that there
was no agreement at all between them; a patient may
be called highly suggestible by one, non-suggestible
by a second, and average by a third. Clearly, whether
or not a patient was rated as 'suggestible' depended
not on any qualities within himself but rather on which
particular psychiatrist happened to carry out the rating
procedure.

This is an example of a fundamental difficulty which is
inherent in all ratings. What the rater says about the
ratee may be taken as evidence about some quality inherent
in the ratee, but it may also be taken as evidence of some
quality inherent in the rater. Two examples may show
what I have in mind. If the same set of schoolboy essays is
rated by two schoolmasters on a ioo-point scale, one may on
the average give a mark of 80, the other a mark of 40. We
would not regard this as showing anything about the quality
of the essays, but would rather say that one schoolmaster
was more severe in his marks than the other. In other words,
we would use his rating to come to a conclusion, not about
the things rated, but about him, the rater. Or let us con-
sider another situation in which a person is asked to rate on
a scale of excellence both Shakespeare's Hamlet and the
latest Hollywood masterpiece. We would not regard his ver-
dict as telling us very much about Shakespeare and his ex-

Can Personality be Measured? 187

cellence, but we would regard it as telling us something
about the person who made the rating.

This difficulty is quite inescapable, of course. We are
throughout dealing with a process of interaction between
two people, and what we have to do is to analyse this inter-
action rather than accept any part of it as being objectively
true. There is ample evidence to show that this statement is,
in fact, correct. Just one such study will be quoted because
of its importance. The hypothesis on which this study was
based may be phrased as follows. An individual judging
others with respect to a given trait will be determined in his
judgement not only by objective reality, but also by his own
possession of this trait. His judgements will be influenced in
two different ways, according to whether he is conscious of
the fact that he himself possesses the trait in question, or
whether he is unconscious of this fact. If he is unconscious of
possessing the trait himself; if, in other words, he lacks in-
sight, then he will tend to attribute a greater amount of that
trait to other people, v/hile, when he has insight into his pos-
session of that trait, he will tend to attribute a lesser amount
of it to other people.

The experiment was run along rather simple lines. A
group of people was chosen, each of whom knew the others
quite well, and they were all asked to rate themselves and
each other on a number of traits. As an example, let us take
the trait of stinginess. First of all, those people were selected
who were considered stingy by the majority; they, in other
words, might be considered to possess this trait to an above-
average degree. They were then divided into two groups, i.e.
those who had insight and rated themselves as being stingy,
and those who lacked insight and rated themselves as not
stingy.

As a next step, the ratings made by these two groups of
judges of all the other people in the group were compared
with the average ratings made by the whole group. The
hypothesis predicted that those judges who were stingy and
had insight would be rather lenient, and not attribute stingi-
ness to other people, while those judges who were stingy and

1 88 Part Two • Personality and Social Life

did not have insight would be rather strict, and attribute
stinginess to other people to a greater extent than objectively
justified. This, indeed, is precisely what happened, and we
can see, therefore, that to understand the meaning of a rat-
ing given by one person of another, we must have some
knowledge, not only of the meaning attributed to the terms
used by the rater, but also to his own possession of the trait
in question and to his insight or lack of insight. This, of
course, enormously complicates the whole problem of
ratings.

Unfortunately, these are not the only difficulties that
arise ; there are many more. One of the most ubiquitous is
the so-called ' halo ' effect, a term used to denote a tendency
common among human judges to like or dislike the ratee as
a whole, and thus attribute all the desirable and admirable
traits to him if we like him, or to attribute all the undesir-
able and not so admirable traits to him if we dislike him.
Thus, again, the relationship of the rater to the ratee in-
fluences the ratings given, and makes it more subjective and
less valid than we would like it to be. It is very difficult to get
around this halo effect unless, perhaps, it could be done by
always having two judges for each one of our ratees, one
who liked and one who disliked him intensely. However,
this is a difficult thing to arrange in practice, and the end re-
sult might be that everyone would receive an average rating
because the two judges would completely cancel out.

A last point which should be mentioned is that there seem
to be very considerable differences in rating ability among
human beings. Some people apparently have an almost un-
canny ability to come up with the right answer, while others
are almost constantly wrong in their ratings and their pre-
dictions based on these ratings. Thus, it was found during
the war that in the War Office Selection Boards, where a
number of judges made predictions as to the ' officer quality5
and future career of young officer candidates, some people
had a batting average of correct predictions very much in
excess of that of their colleagues, while others were quite un-
usually incompetent. This finding has recently been sub-

Can Personality be Measured? 189

jected to experimental tests in the United States along
several different lines. In one experiment the ratee was asked
to fill in a personality questionnaire of the kind to be dis-
cussed presently. Having done this, he was asked to go into
another room, where he had to stand on a platform, smoke
a cigarette, walk up and down, write on a blackboard ' Mary
had a little lamb', and recite a poem. He did this in front of
a fairly large group of people - the 'raters' in this experi-
ment - who observed him carefully and were then required
to fill in the personality questionnaire in the way in which
they thought he himself had filled it in. It thus became pos-
sible to compare the actual answers given by the ratee with
those which the raters thought, from their brief acquaint-
ance, he would have made.

It was found again that some people were very much bet-
ter than others, and this ability to carry out a task of this
kind was apparently a fairly general one, because when the
experiment was repeated several times with different people
taking the part of the ratee, it was the same raters, again and
again, who succeeded in estimating correctly the answers
that had been given to the questions in the personality
inventory.

In a rather more thorough investigation a slightly differ-
ent method was used. Short interviews with several ratees
were recorded on a sound film and were then played to
groups of judges who had to answer questions about the per-
sonality of the interviewees. The answers to these questions,
which dealt partly with verbal behaviour, partly with be-
haviour in various life situations, were known from an ex-
haustive study of each subject's life-history, test perform-
ance, and so forth. In this way, the judgements of the ratees
could be compared with actual fact. Again it was found that
considerable differences existed in the ability of the raters to
judge the personalities of the subjects, and that psychologi-
cal ability appeared to be a prominent characteristic of a
given judge, regardless of which of the several subjects he
was rating.

Important as this psychological ability undoubtedly is,

190 Part Two • Personality and Social Life

very little is, in fact, known about it. As you might have ex-
pected, intelligence and emotional stability tend to be posi-
tively correlated with it, but the relationship is not close
enough to make selection on this basis very fruitful. One
might think that special training in psychology and psychia-
try would be of help, but in actual fact this does not appear
to be so. Students of the natural sciences are usually superior
to other groups in making judgements of this kind. Psychia-
trists have been found to be reasonably accurate in predict-
ing verbal behaviour, but not in predicting behaviour in
actual life situations. This, in a way, is perhaps what one
might have suspected. Practically the only relationship be-
tween the psychiatrist and his patient is one involving words
and verbal behaviour; unfortunately, what a person says
and what a person does are not necessarily the same thing,
and one cannot extrapolate from the one to the other.

Perhaps the superiority of students of the natural sciences
over psychiatrists and clinical psychologists is not to be won-
dered at either. Physicists, chemists, engineers, and so on,
are trained to deal with facts and not to indulge in specula-
tion and complex theorizing unsupported by evidence.
Psychoanalysts and other clinical workers, who have not
undergone this type of training, all too easily take their
highly speculative theories seriously, forgetting the very
small factual basis supporting them, and make unwarranted
and complex generalizations and predictions which are not
in accord with reality. It is not known whether this is the
correct explanation of the findings, but it certainly is a pos-
sible one, and it is to be hoped that experimental evidence
will soon be forthcoming to prove or disprove it. There is,
in fact, a good deal of evidence to show that experts in the
more esoteric fields of clinical psychology tend to be less
accurate in their predictions than beginners, and the reason
for this appears to lie mainly in precisely this tendency for
over-elaboration on insufficient evidence on the part of the
expert.

With such very marked differences apparent between
different judges, it seems reasonable to expect that while

Can Personality be Measured? 191

good judges might produce worth-while results, poor judges
could not be relied on to give ratings of any reasonable de-
gree of usefulness. Unfortunately, good judges are rare and
difficult to find, and the procedure for detecting them is so
complex and cumbersome at the moment, that it has never
in fact been used, except for experimental purposes and in
the laboratory. It might be hoped that in the selection and
training of psychiatrists, psychoanalysts, and clinical psycho-
logists, room would be found for the application of experi-
mental procedures of this type, if only to demonstrate to the
majority the fallibility of human judgements on such com-
plex questions and problems. However, to date all that can
be said is that for the great majority of people the method of
rating is a very unreliable and insecure way of arriving at
accurate measures of personality variables, and that clearly
other methods, if available, would be preferable.

In discussing these other methods, as we shall do pre-
sently, we must bear in mind one great difficulty which
makes this branch of psychology particularly frustrating and
unsatisfactory. This difficulty relates to the almost complete
absence of any reliable criteria against which we can evalu-
ate our methods and tests. In psychology, a test is considered
valid when it can be shown, in fact, to measure what it at-
tempts to measure. For instance, let us suppose that we have
been given the task of preparing a battery of tests for the im-
proved selection of sagger-makers' bottom knockers. While,
technically, this may not be an easy task, in principle it is
capable of a relatively straightforward solution, because we
have a criterion against which to measure the adequacy of
any test which might be suggested. We could give our test,
or battery of tests, to 1,000 sagger-makers' bottom knockers,
record their scores, and then correlate these scores with the
actual number of bottoms knocked per day, or per week, or
per year, by each one of our subjects. Once these records
had been made available by the sagger-makers, the relative
capacity of their bottom knockers could be established, and
the relationship of this ability to the test scores would be
known. We could then retain those tests which showed a

1 92 Part Two • Personality and Social Life

high relationship with bottom-knocking ability, and elimin-
ate those of low or moderate predictive ability.

There would, of course, be all sorts of practical difficulties.
Thus, London bottoms and Glasgow bottoms might not be
of exactly the same size, or again, the instruments used
might not be identical. Apprenticeship schools might differ,
and there might be an overall agreement among bottom
knockers not to produce more than a certain number of
specimens each day. While practical difficulties of this kind
certainly make the life of the industrial psychologist more
adventurous and troublesome than it would otherwise be,
they are not insuperable, and ways and means can usually
be found to take into account troublesome factors of this
kind.

The difficulty in the personality field is that for those
variables in which we are really interested there is no outside
criterion of any kind. If there was some certain way of know-
ing who had and who did not have a good sense of humour,
then we could easily correlate any judgements used against
the true facts and find out in this way who was and who was
not a good judge. In a similar way we could construct tests
of ' sense of humour ', retaining those which correlated highly
with our criterion, and throwing out those which did not.
However, if we did have such a perfect criterion, then, of
course, we would have no need of either ratings or tests; the
fact that both ratings and tests are still used indicates more
clearly than anything that no objective and valid criterion is
in practice available.

The extraordinary difficulty of finding a good criterion
may be illustrated by certain experiences gained during the
war, when the problem of pilot selection became a crucial
one. There are, quite obviously, considerable differences be-
tween people in their ability to learn to fly and to pilot a
plane. It would be a very inefficient and expensive method
to train everybody who applied, and to count those as fail-
ures who smashed up a plane, which might have cost half a
million pounds. Consequently, a number of objective tests,
of the kind to be described presently, were constructed be-

Can Personality be Measured? 193

cause it was thought that the abilities measured by these
tests would be relevant to the art of flying a plane. However,
no relationship whatever was observed between scores on
these tests and actual flying ability. This seemed unreason-
able, because much previous work had shown that such a
relationship did, in fact, exist, and consequently the criterion
of flying ability became suspect. This criterion, which on its
face appeared quite a reasonable one, consisted in the rating
of an experienced instructor, who carefully observed the
pupil going through a series of predetermined manoeuvres;
each manoeuvre was rated separately as to the excellence of
its execution, and the instructor then gave an over-all estim-
ation of the number of points gained in these various
manoeuvres and his over-all impression of the candidate in
flight.

In order to check up on the reliability of this criterion a
number of candidates were asked to go through the set of
manoeuvres, watched this time by several independent in-
structors who made their judgements without consulting
with each other. When their ratings were compared, it was
found that there was hardly any agreement between them at
all. They all marked down a candidate who broke off the
under-carriage of his plane in an attempt to land, up-ended
the plane, and finally managed to break its back. However,
this candidate constituted the solitary case on which there
was universal agreement. For the rest, the subjective
opinions of the judges were so divergent that, quite clearly,
no reasonable criterion could be expected from unreliable
ratings of this type. It took a very long time indeed before it
became possible to objectivize the procedure.

The main alternative method to that of rating has tradi-
tionally been the questionnaire or personality inventory. In
this, the subject in whom we are interested is asked to an-
swer a number of questions about himself; from his answers
to these questions we try to arrive at some conclusions
about his personality. This method obviously has a number
of drawbacks and difficulties to which we shall turn pre-
sently. These, however, have not prevented it from working

194 Part Two • Personality and Social Life

very efficiently in certain situations, and of all the methods
of collecting data of relevance to personality study in psycho-
logy, the questionnaire has probably been the most widely
used.

It originated very much as did intelligence tests, in re-
sponse to a practical emergency. During the First World
War the American army was getting worried over the large
number of recruits who were boarded out because of neuro-
tic disorders of one kind or another. To the army this was a
serious matter. It meant, for one thing, that large numbers
of medical personnel, nurses, orderlies, and so on, had to be
employed to look after these neurotics, special hospitals had
to be set up, psychiatrists had to be employed to determine
the degree of incapacity, and last, but not least, a consider-
able financial sacrifice was involved in boarding out the men
under this category. It has been calculated that it cost the
United States Government 75,000 dollars in all for just one
neurotic to be 'separated5 from the army, as this somewhat
painful process is called in the United States. When 75,000
dollars are multiplied by hundreds of thousands of cases, it
will be clear that it would be of tremendous help to the
authorities to be able to diagnose the potential neurotic be-
fore he ever joined the army. Such early diagnoses would be
exceedingly helpful, not only to the army, but probably even
more to the potential neurotic. It would prevent him from
going forward to an almost certain nervous breakdown, and
he would probably be far more useful to the country doing
a civilian job involving little stress and not precipitating all
the neurotic symptoms with which we have become so fami-
liar in the last fifty years, than as an emotional victim of
army life and hazards. From the ethical point of view, this
argument has sometimes been attacked because it means
that the burden shouldered by different individuals is not
the same, but whatever the ethical implications may be, few
armies would, for practical reasons alone, refuse to employ
a selection procedure which would 'screen out the neurotic',
if such screening were possible.

The person to whom the difficult task of creating such a

Can Personality be Measured? 195

sieve was entrusted was Woodworth, one of the most emi-
nent of American psychologists. Realizing that laboratory ex-
periments would not be feasible under army selection con-
ditions, Woodworth drew up a questionnaire which recruits
had to answer. This questionnaire consisted of very simple
questions which had to be answered by underlining 'Yes',
'No', or, where the answer was difficult, by underlining a
'?'. The questions themselves were obtained by going
through several text-books on psychiatry and noting all the
symptoms associated with neurosis, such as having many
headaches, having nightmares, worrying about one's health,
often feeling miserable, lacking self-confidence, having feel-
ings of inferiority, or worrying too long over humiliating ex-
periences. Woodworth put together over 200 questions of
this kind in his final inventory, but it was never actually em-
ployed for selection, as the war had come to an end by then.
The reader who is interested in the kinds of question asked
may read through a very much briefer questionnaire used in
GreatBritain, which is printed below. It consists of forty ques-
tions, and the score is simply the number of 'Yes's' under-
lined by the subject. The average score of groups of normal
subjects is about 10, that of neurotics about 20. (The reader
should not take it too much to heart if, on going through the
questionnaire, he finds that he gives 20 or more 'Yes' an-
swers. This does not mean that he is a neurotic. Inventories
of this kind can give useful leads to the experienced psycho-
logist and they may single out people for further investiga-
tion; in themselves they should never in any circumstances
be used to arrive at a conclusion about a person's mental
health.)

MAUDSLEY MEDICAL QUESTIONNAIRE

Read through these questions and underline the correct answer,
either 'Yes' or 'No'. Do not omit any item. It is important that
you should be quite frank.

(1) Do you have dizzy turns? Yes No

(2) Do you get palpitations or thumping in your

heart? Yes No

196 Part Two • Personality and Social Life

(3) Did you ever have a nervous breakdown? Yes No

(4) Have you ever been off work through sickness a

good deal? Yes No

(5) Did you often use to get 'stage fright' in your life? Yes No

(6) Do you find it difficult to get into conversation

with strangers? Yes No

(7) Have you ever been troubled by a stammer or

stutter? Yes No

(8) Have you ever been made unconscious for two

hours or more by an accident or blow? Yes No

(9) Do you worry too long over humiliating experi-

ences

Yes

No

Yes

" No

Yes

No

10) Do you consider yourself rather a nervous person?

11) Are your feelings easily hurt?

12) Do you usually keep in the background on social
occasions? Yes No

13) Are you subject to attacks of shaking or trem-
bling?

14) Are you an irritable person?

15) Do ideas run through your head so that you can-
not sleep?

16) Do you worry over possible misfortunes?

17) Are you rather shy?

18) Do you sometimes feel happy, sometimes de-
pressed, without any apparent reason?

19) Do you daydream a lot?

20) Do you seem to have less life about you than
others?

21) Do you sometimes get a pain over your heart?

22) Do you have nightmares?

23) Do you worry about your health?

24) Have you sometimes walked in your sleep?

25) Do you sweat a great deal without exercise?

26) Do you find it difficult to make friends?

27) Does your mind often wander badly, so that you
lose track of what you are doing?

28) Are you touchy on various subjects?

29) Do you often feel disgruntled?

30) Do you often feel just miserable?

3 1 ) Do you often feel self-conscious in the presence of
your superiors?

(32) Do you suffer from sleeplessness?

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Yes

No

Can Personality be Measured? 197

(33) Did you ever get short of breath without having

done heavy work? Yes No

(34) Do you ever suffer from severe headaches? Yes No

(35) Do you suffer from * nerves'? Yes No

(36) Are you troubled by aches and pains? Yes No

(37) Do you get nervous in places such as lifts, trains,

or tunnels? Yes No

(38) Do you suffer from attacks of diarrhoea? Yes No

(39) Do you lack self-confidence? Yes No

(40) Are you troubled with feelings of inferiority? Yes No

In view of the great ease with which questionnaires can be
drawn up and the large number of people who can be tested
simultaneously without difficulty, questionnaires became
very popular indeed in the years between the wars. They pro-
vided so much grist for the statistical mill of the psycho-
logist's calculating machines that few people managed to ask
critical questions about the meaningfulness of data so ac-
quired. Gradually, however, doubts began to grow, even in
the most enthusiastic. Many people had experiences of a
kind to make them somewhat suspicious of the truthfulness
of the answers given by subjects. Thus, one might see some
unfortunate individual sitting down with his questionnaire,
his hands trembling and sweating with excitement, his face
getting pale and flushed alternately, and his tongue licking
his lips, his whole body in a tremor of nervousness ; on going
over to reassure him, one would find that after the question,
'Are you generally a nervous sort of person?' he had boldly
put the answer ' No ' !

Doubts of this kind came to a head in the 1930s because
some very odd results indeed came out of the statistical mill.
To explain the consternation caused, we must go back a
little to the 1920s again. While following Woodworth, most
questionnaires had dealt with the diagnosis of neurotic dis-
orders and emotional instability, the translation of Jung's
book on Psychological Types made people interested in the
investigation of extraversion and introversion, and many
questionnaires were drawn up for this purpose. The pro-
cedure followed was pretty much the same as that used by

198 Part Two • Personality and Social Life

Woodworth. Jung's book would be gone through with a
fine-tooth comb and statements about the behaviour of the
typical extravert or introvert could then be transformed into
questions to be answered 'Yes' or 'No'. In this way, long
and detailed questionnaires and inventories were drawn up
for the measurement of a temperamental trait which, ac-
cording to Jung's hypothesis, was completely different and
unrelated to neuroticism.

For about ten years psychologists went along their way
happily measuring neuroticism and introversion with these
different instruments, when suddenly someone hit upon the
bright idea of comparing the scores made by the same group
of people on different types of questionnaires. Now, if all the
neuroticism inventories measured neuroticism, presumably
the same people, i.e. the emotionally unstable ones, should
have high scores on all the questionnaires, while another
group of people, i.e. the emotionally stable ones, should
have low scores on all the questionnaires. Similarly, with re-
spect to the introversion questionnaires, all the people who
were introverted should have high scores on all the question-
naires, while the extraverted ones should have low scores on
all the questionnaires. And as neuroticism and introversion
were supposed to be quite unrelated to each other, the fact
that a person had a high or a low score on neuroticism
should not in any way determine his score on an introversion
questionnaire.

The many investigations conducted to test these hypo-
theses all arrived at the same conclusion. The relation be-
tween one neuroticism inventory and another was not at all
close; the relationship between one introversion measured
and another was not at all close ; worst of all, the relation-
ship between a measure of introversion and a measure of
neuroticism was just about as close as that between two
measures of introversion or two measures of neuroticism! In
other words, we are dealing with haphazardly grouped
items, or questions, which do not succeed in measuring any
known or hypothesized form of behaviour, and which, be-
cause of their arbitrary selection, cannot even be used to test

Can Personality be Measured? 199

any kind of worth-while hypothesis. Questionnaires of this
type, therefore, are useless, misleading, and must be re-
garded as quite fatal to the development of the scientific
study of personality. It is small wonder, then, that question-
naires fell very low indeed in the estimation of psychologists
and that it needed a complete revolution in their construc-
tion before they were readmitted to the armamentarium of
the student of personality.

The reassessment began by noting that one very funda-
mental assumption had been made by questionnaire con-
structors which had little basis in fact. The assumption was
that a person in answering a question would be giving an
objectively true answer. Now, there are many reasons why
this is not only unlikely, but impossible. Let us take such a
question as 'Do you have frequent headaches?' Let us
assume that the person who is filling in the questionnaire is
willing to give a truthful answer. What shall he say? Does
one headache a week qualify as being frequent, or should it
be more than two a week, or perhaps one every day? How
severe ought a headache to be in order to qualify? Perhaps
the question means 'more frequently than the average' ; but
what is the average? How many headaches do people have
in the course of a year? It is clear that the answer is subject
not only to the actual number of headaches which a person
has and their severity, but also to his estimate of what the
questioner had in mind, his estimate of what the average
number of headaches is in the community, and various
other factors as well. How can we then attribute any direct
meaning to the 'Yes' or the 'No' which is the only answer
permitted to the subject? If we regard this reply as a truth-
ful account of his own experience, we must come to the con-
clusion that it is quite impossible to interpret in any reason-
able manner. If the reader will go through the forty ques-
tions in our questionnaire, he will find that nearly all of
them are of a similar kind and subject to a similar criticism.
How easily do one's feelings have to be hurt in order to an-
swer ' Yes ' to question 1 1 ? How difficult does it have to be
for one to get into a conversation with strangers before one

200 Part Two • Personality and Social Life

answers ' Yes ' to question 6 ? How shy does one have to be
to say 'Yes' to question 17, and how worried about one's
health to say ' Yes ' to question 23 ? It will be obvious that the
whole basis of the questionnaire, as it used to be conceived,
is at fault.

Another difficulty which arises is that a person may not
know the truth about himself and may therefore not be able
to give a correct answer. As we have seen before, a stingy
person may be quite unconscious of his stinginess and com-
pletely lack insight. Under those conditions we can hardly
expect him to give us an answer which will be very revealing
or diagnostic of the true state of affairs. Few people, for in-
stance, who have a poor sense of humour are conscious of
this fact. In a well-known experiment the question was put,
'Do you have a better sense of humour than average?'
Ninety-eight per cent of those who answered the question
claimed to have a better than average sense of humour !

A third point, which has always created difficulties in
accepting questionnaire responses at face value, lies in the
fact that these are so easy to fake. Most people want to be
seen in the best light; consequently, they tend to 'put their
best foot forward', or in other words to deny those traits and
those behaviour patterns which are socially regarded as be-
ing unreliable, immature, unstable, and so forth. Why
should we expect a person to indict himself in a question-
naire; what right have we to expect him to tell the truth
about himself if that truth is uncomplimentary and would
show him up as a rather poor sort of fish? We have no
guarantee at all, then, that a person would even attempt to
tell the truth, even assuming that he did know the truth and
that he could make any sense out of our questions.

In recent years the whole basis of interpretation of ques-
tionnaires has shifted, and a serious attempt has been made
to get over these difficulties. In the first place, psychologists
have more or less abandoned the idea that the answer a per-
son gives should be intrepreted as a truthful self-revelation.
We are now concerned, not at all with the interpretation of
answers, but merely with the objective fact that a person

Can Personality be Measured? 201

puts a mark in one part of the paper rather than in another.
Thus, it is a fact that a person underlines the 'Yes' or the
'No' after the question 'Are you touchy on various sub-
jects?'; it is a subjective interpretation of this objective fact
to assume that he has answered it in a truthful manner and
that in fact he is touchy on various subjects. To what use, it
may be asked, is the objective fact of a given endorsement if
we cannot interpret it? The answer is a very simple one in-
deed. Let us take two groups of people between whom we
want to differentiate. For the purpose of illustration, let us
take a group of normals and a group of neurotics. We might,
of course, have taken a group of introverts and a group of
extraverts, or a group of epileptics and a group of patients
with brain damage. The procedure is quite general, and can
be applied whenever two criterion groups can be distin-
guished.

Let us now apply our questionnaire to a group of 1,000
normals and 1,000 neurotics, and let us note merely the ob-
jective fact of how many in each group underlined the 'Yes*
and how many underlined the 'No' alternative printed after
each question. Let us take the question, ' Do you suffer from
sleeplessness?' It has been found that 32 per cent of neuro-
tics answer this in the affirmative, whereas only 13 per cent
of normals answer it in the affirmative. Now, this is an ob-
jective fact. We are not here concerned with the reasons why
a greater number of neurotics endorse this question. It might
be that neurotics are truly more given to sleeplessness than
are normal people. It could be that neurotics are more given
to complaining, although, objectively, normal people suffer
more from sleeplessness. It could be that both normals and
neurotics suffer equally from sleeplessness, but that this pre-
sents a much greater stress to the emotionally unstable
neurotic. Whatever may be the true reason, the fact remains
that neurotics and normals are differentiated objectively in
their behaviour, and that consequently, when dealing with
people whom we want to classify into either of these groups,
the endorsement of the ' Yes ' answer to this question makes
it more likely that they belong in the neurotic group than

202

Part Two • Personality and Social Life

that they belong in the normal group. Knowing the per-
centage endorsements of known members of the two groups,
we can calculate exactly the probability of a given person
belonging to either of the two groups from knowing his
answer to this particular question.

Figure 4 shows the existing differences in endorsements of
normal and neurotic groups of 1,000 each on sixteen items
in the Maudsley Medical Questionnaire. In making a judge-
ment one would, of course, not rely on the answers to single
questions alone, but probabilities can be combined, and by
the time we have combined probabilities on forty questions

£H| Neurotics
\ I Normals

1 2 3 10 11 13 14 16 22 32 33 35 36 37 '39 40

Figure 4: Percentage of normal and neurotic subjects answering 'Yes'
to selected questions from the neuroticism inventory given on page 195.

we can come to a fairly reasonable estimate as to whether a
person belongs in the neurotic group, in the normal group,
or whether his status is indeterminate, being somewhere in
between the two. We would, of course, want to make certain
of the validity of the argument by following up groups with
high and low scores, but, by and large, we may feel reason-
ably content that in changing our basis of interpretation we
have left behind the morass of unwarranted interpretation
and are now on fairly firm ground.

We also have a method now which enables us to eliminate
poor items and retain good items. Any item which does not
succeed in differentiating between normal and neurotic

Can Personality be Measured? 203

groups would, on these grounds, be a poor item, however
prominent it might be in the psychiatric text-books, while
any item which did give a good differentiation between a
normal and a neurotic group may be a good item, although
it may not even be mentioned in the text-books. Our ap-
proach is a purely empirical one, and this enables us to call
to our aid complex statistical and mathematical analyses.
Thus, brief and simple questionnaires like the one presented
on an earlier page can be constructed on an empirical basis
and are, in fact, more valid and more reliable than the very
much longer ones current in the 1920s.

But how can we get over the difficulty of faking? How is it
possible to prevent our subjects from answering all the ques-
tions in the positive 'healthy', non-neurotic direction? The
answer is that, while it is not possible to do this, it is possible
to detect tendencies of this kind and either to make allow-
ances for them, or to disregard the result of the question-
naire. The method used, that of the so-called 'lie-scale', is
essentially a very simple one. A person who wants to give a
particularly good impression of himself does so by answering
'No' to all questions which would put him in a somewhat
less good light than that in which he wants to be seen. Con-
sequently, all we have to do to catch him is to make up a
scale of items which, while they put people in a rather
poor light, are nevertheless of such a nature that anybody
would have to admit to the answer ' Yes ' if he was at all
honest. As an example we might take a question like the fol-
lowing: 'Have you ever told a white lie?' Certainly, to have
to admit to having told lies is to put oneself in a poor light,
yet very few people could truthfully answer ' No'. If a person
does answer ' No ' to this question, and to a large number of
others of a similar kind, then we can be fairly certain that
his questionnaire answers also are not to be trusted. It is, of
course, conceivable that a person might, quite truthfully,
answer 'No' to all the questions on the lie scale, but in that
case he would be a saint, and questionnaires are not con-
structed to deal with saints.

Another method which has very much improved the

204 Part Two • Personality and Social Life

accuracy of questionnaire measurement has been that of de-
tailed internal statistical analysis. This analysis involves the
calculus of implication, which is discussed in more detail in
the chapter on 'Personality and Polities'. Briefly, what this
means is simply that the psychologist analyses the relation-
ships between answers to different questions. Logically,
questions such as 'Are your feelings easily hurt?5 and 'Do
ideas run through your head so that you cannot sleep?' are
quite independent; it would be possible to answer 'Yes' to
one and 'No' to another. In actual fact, people who answer
'Yes' to one tend to answer 'Yes' to the other, and vice
versa. In other words, there is a factual implication, so that
from knowing a person's answer to the one we can predict
his answer to the other. It is possible to give a numerical
statement of the degree of implication, and the statistical
psychologist has worked out methods to analyse the net-
work of implications running through all the questions in a
questionnaire. The methods used are too technical to dis-
cuss in detail here, but they have helped enormously in
making questionnaires more reliable, in cutting out dead
wood, and generally in improving measurement along these
lines.

Modern questionnaires, then, are usually quite different
in their method of construction and interpretation from
those earlier ones which encountered such severe criticism.
A great deal of work has usually gone into their construction
as compared with the a priori, happy-go-lucky methods
which were responsible for the original neuroticism and
introversion inventories. Detailed experimental studies with
different diagnostic groups are usually carried out, and a
complex system of statistical analysis along the lines of the
calculus of implication has to be gone through. The result-
ing instruments have considerable usefulness from the prac-
tical point of view, and may aid the psychologist in his
attempts to measure personality.

In certain ways there undoubtedly are similarities be-
tween ratings and questionnaires. Thus, the psychiatrist who
wants to rate a person's degree of emotional instability, or

Can Personality be Measured? 205

neuroticism, will ask him a series of questions in an inter-
view, which is essentially very similar to the questions con-
tained in the questionnaire. Indeed, experiments have
shown that there is considerable agreement between the
questionnaire replies given by large samples of people, and
the ratings of the same people made by a highly skilled
psychiatrist who interviewed them. If the outcome is so simi-
lar, why, it may be asked, do we feel more confidence in the
questionnaire than in the rating procedure?

There are several reasons for this. In the first place, a great
deal of work has gone into the exact choice and wording of
questions in the questionnaire. The forty items in the Mauds-
ley Medical Questionnaire are the remains of almost 1 ,000
which were tried out, most of which were discarded
when it was experimentally found that they did not dis-
criminate as well as those which were retained. The psychia-
trist who interviews a patient does not usually perform the
considerable amount of work which would be involved in
the careful selection of questions on an objective basis. Con-
sequently, his choice is almost inevitably inferior to that
leading to the construction of a questionnaire.

In the second place, data, once they have been gathered,
have to be interpreted and compared. This is best done by
statistical procedures relating the scores obtained by a given
person to replies given by large control groups whose re-
sponses are known. The psychiatrist seldom asks exactly the
same questions of any two people, and he never keeps exact
statistics to tell him what the means and averages of differ-
ent groups are. Consequently, he does implicitly and vaguely
in his head what the psychologist does explicitly and with
the use of calculating machines.

In the third place, the psychologist using an inventory or
questionnaire is careful to collect large groups of normal
people, which will give him a base-line in terms of which to
make his interpretation. The psychiatrist hardly ever sees
normal people in the course of his work and therefore lacks
this kind of objective basis.

An actual experiment may show the quite remarkable

206 Part Two • Personality and Social Life

differences in judgement which these various sources of
errors may produce among psychiatrists. These figures re-
late to the medical examination in the United States of all
literate recruits to the United States Army, prior to their in-
duction. This work was carried out during August 1945 in
some fifty stations all over the country. Fourteen per cent of
these candidates were rejected for psychiatric reasons, and
it is very instructive to analyse the reasons given for rejec-
tion in the various induction stations. There is ample evi-
dence to show that candidates going to these various stations
did not differ very much from each other, so that differences
in diagnosis cannot be due to the candidates themselves, but
must be due to the examining psychiatrists.

First of all, then, let us consider the percentage of psychia-
tric rejections in these various induction stations. As men-
tioned above, the overall percentage is 14. In some induc-
tion stations only -5 per cent, i.e. only one person in 200, was
rejected for psychiatric reasons. At another induction station
the percentage was 50-6, i.e. every other candidate was re-
jected for psychiatric reasons. To put it slightly differently,
for every one inductee rejected for psychiatric reasons in one
centre, there were over 100 rejected in another!

When we look at the reasons why people were rejected we
find again considerable variation from one centre to an-
other. Of all those rejected for psychiatric reasons, 40 per
cent on the average were diagnosed as * psychoneurotic5.
However, this average concealed a very large amount of
variability. Thus, in one centre 27 per cent were diagnosed
as psychoneurotic; in another, 90-2 per cent. Differences in
other psychiatric categories were equally large, or larger,
and it must be quite clear that ratings of this type are sub-
ject to all the sources of error mentioned in the preceding
paragraph.

In contrast, questionnaires given at various centres to
similar groups of candidates did not disclose any such varia-
bility because it was now possible to objectify the selection
method, to set definite standards, and thus to gain all the
advantages of an objective as opposed to a subjective ap-

Can Personality be Measured? 207

proach. This is only one example of many which could be
given, but it does illustrate that, although in principle there
may not be very much difference between the procedure of
interviewing and rating, on the one hand, and the procedure
of administering questionnaires, on the other, yet there are
very great advantages attached to the latter procedure
which should not be ignored.

We must now turn to a third method of investigation
which is very much more important and scientifically fruit-
ful than those considered hitherto, and which will become
more and more the standard method of personality investi-
gation. Here, what we are using is not what one person says
about another, or what a person says about himself, but
rather we make an actual measurement of the observed be-
haviour of a person in a situation or a test so constructed as
to call forth the type of behaviour in which we are inter-
ested. It will be easiest to explain this type of measurement
by including in the discussion a consideration of a theoretical
point mentioned at the beginning of this chapter, but not
discussed in full. We noted at the beginning that the man in
the street discusses personality variables in terms of such
concepts as 'traits', and we have followed his example so far
without going in any detail into the very important question
of whether there is any evidence for the existence of such
traits, and what precisely we understand by this whole
notion. Let us remedy this omission and begin by consider-
ing a given trait, say that of persistence.

What do we mean when we say that a given person is per-
sistent? We mean, surely, that he tends to carry on an activ-
ity in the face of boredom, or pain, or exhaustion, long after
other people less persistent than himself might have stopped.
Implied in the notion of this trait of persistence is the assump-
tion that persistence will be shown by a given person, not
only in one particular situation, but in a wide range of situa-
tions. We would not call a person persistent if he only con-
tinued one particular type of activity against fatigue and
boredom, but let himself be put off continuing other activi-
ties. Thus, the notion of generality is quite essential to our

208 Part Two • Personality and Social Life

concept of a trait; a trait is something which is shown in a
large number of different situations.

The difficulty caused by this requirement is dimly noted
in ordinary language too. Thus, we have a concept of
* cowardice' as a trait, but quite obviously this does not work
too well. The V.C. may faint in the dentist's chair; in other
words, he may be a coward in one situation and a hero in
another. The schoolboy may be physically brave, but may
not dare to show that rare quality of moral courage. It be-
comes a factual question to find out whether a given hypo-
thetical trait is, in fact, general in the sense of being dis-
played in a wide variety of situations, or whether, in fact, all
we are dealing with is a specific response to a specific situa-
tion.

How would this apply to the problem of the measurement
of persistence? What we would do here would involve a
whole series of steps which may be taken as exemplifying the
method employed by the modern psychologist in measuring
a particular trait of personality. In the first place, we would
make a theoretical statement of the nature of the trait under
investigation somewhat along the lines indicated in the pre-
ceding paragraph, where we defined persistence as the ten-
dency to carry on an activity, in spite of fatigue, pain, or
boredom. The second step would then be the construction of
a large number of tests or situations making possible for the
subject the demonstration of different degrees of the trait in
question. Here are a few such tests, quoted more or less at
random from the very large literature that has grown up
around the concept of persistence. Most of these have been
used in connexion with schoolchildren, but many of them
have been adapted for use with adults also, and the results
have been just as favourable.

Here, then, is our first test. The subject is given a large
jig-saw puzzle which he is told to fit together. He is told that
this is a test of intelligence, or some other kind of ability, and
that he has as much time as he wants in order to do the test.
Before handing the jigsaw puzzle to the subject, however,
the experimenter has removed some of the pieces and has

Can Personality be Measured? 209

substituted others for them in such a way that it is quite im-
possible to complete the test successfully. (This, of course, is
done in such a way that the candidate has no idea of what
has happened and cannot, in fact, guess at any stage that his
task is objectively impossible.)

The subject begins to put the test together, gradually gets
tired as he fails to advance beyond a certain point, and
finally is discouraged and gives up completely, saying 'This
is too difficult for me', or 'I don't think I'll be able to do
this'. In extreme cases, subjects have even been known to
throw the test out of the window, hurl it at the experimenter,
or burst out crying and leave the room in tears. However
they may terminate the experiment, the time from the be-
ginning to the end is taken as their score, because this is the
period during which they have continued with the task.

Now let us look at the next test. The subject here is asked
to pull a dynamometer as hard as he can. (This is an instru-
ment measuring the strength of a patient's pull against a
steel spring, the result in pounds being registered on a large
dial.) Having found the strongest pull of which the subject
is capable, the experimenter then instructs him to pull the
dynamometer at exactly half that strength for as long as he
can. The length of time during which he is capable of doing
this is then taken as another measure of his persistence (not
of his strength - this variable is controlled by making the
tasks relative to the initial pull).

A third test may be given consisting of one of those
Chinese puzzles in which eight carved pieces of wood have
to be put together in the form of a cross. The subject is
shown the completed cross, which is then taken to pieces,
and the test proper begins. While, theoretically, the candi-
date may succeed in doing this, the task is so difficult that,
in practice, no one has managed it before giving up. Again
the length of time taken before giving up is a measure of the
candidate's persistence.

A fourth test might involve an item in an intelligence test
which was well above the candidate's ability to solve. All
items in the test are timed separately, and the time which

210 Part Two • Personality and Social Life

the candidate is willing to spend on an item too difficult for
him to solve is a measure of his persistence.

Hundreds of other similar tests can be used, and have
been used, but the reader will be able to see the general
underlying principle without an unnecessary duplication of
instances. Instead, we will go on to the third stage of the
measurement of persistence. If our hypothesis that persist-
ence is a general trait is correct, and if these are all measures
of persistence, then it must follow that the person who is per-
sistent in any one of these tests should be persistent in the
others, and that a person who lacks persistence in one of
these tests should also lack persistence in the others. Conse-
quently, what we must do is to administer a whole series of
thirty or so tests to a group of subjects, give each subject
a score on each of the tests, and then investigate statisti-
cally the degree to which success in one test predicts
success in all the others, and failure in one test failure in all
the others.

There is ample evidence to show that agreement between
different tests exists to a quite considerable extent. It might
have been thought, a priori, that the intellectual child might
be persistent with a difficult intelligence test item, say, but
lack persistence in the dynamometer pull, while, conversely,
the athletic child might be persistent in connexion with the
dynamometer, but lack persistence on more intellectual
tasks. Up to a point, this is true; there is a tendency for in-
tellectual tasks to hang together, or for physical tasks to hang
together more closely than they do from one group to the
other, but the difference is slight and the over-all agreement
considerable. Consequently, it appears that we can justifiably
talk about a trait of persistence as being manifested in all these
tests. Having shown this, we can then conclude that the best
measure of persistence is a person's average performance on a
well-selected battery of tests of this kind. By well-selected in
this connexion we mean that the tests should cover as many
different areas of ability and interest as possible; that they
should be adequate for the age groups for which they are
designed; that they should not take too long to administer;

Can Personality be Measured? 2 1 1

and that they should not require too much complex
apparatus.

Having designed our battery of tests for the measurement
of persistence, we must next demonstrate that these tests
actually do measure persistence rather than something else.
It might be, for instance, that these are all measures of in-
telligence, and that the intelligent child tends to do better
than the stupid one. This may not seem likely on a priori
grounds, but psychologists are justly suspicious of a priori
grounds and prefer some empirical evidence before coming
to a conclusion. One possible way of answering a question of
this kind is to administer intelligence tests to the same chil-
dren who have been given the persistence tests; it is possible
then to show statistically that whatever it is that the persist-
ence tests have in common, it is not intelligence. Another
method of showing the validity of our test battery in a posi-
tive rather than in a negative way is to find some reasonable
outside criterion and correlate the test results with it. It will
be clear from our discussion of outside criteria that we can-
not be very hopeful of finding a very good criterion, or one
giving high correlations with our tests, but we would never-
theless be rather suspicious of our results if we found that in
correlating them, say, with teachers' estimates of persistence
there was no correlation at all. Equally, we would be rather
suspicious if estimates of a child's persistence given by his
classmates did not correlate at all with his performance on
the test. In actual fact, reasonably high correlations are
found, so that there seems to be little doubt that the test
battery measures more reliably and more validly what
schoolteachers and classmates indicate by their ratings.

A third method which is also often used would be to make
certain deductions from the hypothesis that what we are
measuring is persistence, and then go on to test the truth of
these deductions. Thus, it seems reasonable to suppose that,
given equal intelligence, a persistent person would do better
at school or university than a person lacking in persistence.
Experimentally, this would lead us to apply our battery of
persistence tests to a group of schoolchildren, or students,

212 Part Two • Personality and Social Life

and pick out a set of highly persistent, as well as a set of non-
persistent, subjects. We would then administer intelligence
tests to these groups and equate the persistent and non-
persistent ones in intelligence by eliminating from the more
intelligent group one or two of the most intelligent, until the
groups were equal in I.Q,. Our prediction now would be
that the more persistent group, although no more intelligent
than the non-persistent group, should do better in their
school work, or obtain better degrees at university, than the
non-persistent group. There is ample evidence to show that
this is in fact so, and that persistence contributes noticeably
to success. We may thus rest at this point in the secure con-
viction that we have succeeded in measuring, with a given
amount of reliability and validity, the trait of persistence,
and that we have also succeeded in adducing evidence to
show that this trait is not a mere artefact of the imagination.
I have given as an example a trait where the outcome of
the set of experiments performed by many authors has been
very positive indeed. As the next example, I will give a trait
where the outcome has been very much more complex, and
where the notion of a unitary trait has, in fact, not been
borne out by the evidence. This is the trait of suggestibility,
and the investigations carried out here have an important
lesson to teach. First of all, let us start by noting roughly
what is meant by suggestibility. It is a state or trait causing a
person to perform certain acts without having any motiva-
tion of his own to do so, or even in spite of having a certain
degree of motivation to the contrary. Psychologists have,
for many years, been constructing tests of suggestibility as so
conceived, and a few of these may be mentioned :

(i) Chevreul Pendulum Test. The experimenter draws a
chalk line across the table. In his hand he holds a little pen-
dulum made of a small piece of metal, suspended from a
thread, which he holds in his hand. He tells the subject that
the magnetic lines of the earth go along the chalk line he has
drawn on the table, and that they act on the pendulum,
making it swing to and fro. As he says so he slowly causes the
pendulum to swing along the line on the table. He then

Can Personality be Measured? 213

hands it on to the subject, telling him that he should try to
hold it quite still and steady over the centre part of the line
on the table, but that, in fact, he will be unable to do so. The
magnetic lines of the earth will cause the pendulum to sway
to and fro, to and fro, to and fro. In actual fact, many sub-
jects accept this suggestion and, quite unwittingly and un-
consciously, move their arm in such a way that the pendu-
lum begins to swing along the line suggested.

(2) Body Sway Test. In this, the subject is told to stand
quite still and relaxed with his eyes closed, his heels together,
and his arms by his side. He is told specifically that he
should go on standing like that while a record is being
played to him. This record endlessly repeats the suggestion:
'You are falling; you are falling forward; you are falling
forward; you are falling; you are falling forward now. ...'
The actual amount of sway produced by the suggestion is
measured by running a thread from the collar of the subject
over a series of wheels on to a kymograph marker, which
produces a graphic record of the forward and backward
sway of the subject on a piece of paper, which is being pulled
along by a motor over the surface of a metal drum. Again,
some subjects accept the suggestion and sway to a considerable
extent; some even fall outright and they have to be caught
by the experimenter, who is standing in front of them.

(3) Arm Levitation Test. In this test the subject stands with
his eyes closed and his right arm raised sideways to shoulder
level. He is told to keep it there while a series of suggestions
is played over to him on the gramophone to the effect that
his arm is getting lighter, that it is rising, and so forth. Again,
a record is made of the number of inches that the arm
moves in an upward or downward direction.

(4) Odour Suggestion Test. In this test the subject is told
that he is going to be subjected to a test of his smell sensi-
tivity. He is seated in one corner of the room, while the ex-
perimenter goes to the other corner of the room, picks up a
bottle containing attar of roses. The bottle is uncorked and
the experimenter slowly carries it across to the subject, tell-
ing him to say 'Now' the moment he gets the faintest sniff

214 Part Two ' Personality and Social Life

of the perfume. When the subject says 'Now', the experi-
menter ostentatiously takes up a measuring tape and mea-
sures the exact distance between the bottle and the tip of the
subject's nose. This is then entered into a big red book, and
the experiment repeated with oil of cloves and camphorated
oil. The fourth time, however, the bottle the experimenter
picks up contains nothing but distilled water. He again car-
ries it across to the subject, the implied suggestion of the
whole procedure being that the bottle again contains some
kind of perfume. Many subjects accept this suggestion and
call out 'Now' when the experimenter reaches the same
point where they previously recognized the smell.

(5) Progressive Weights Test. In this test the subject sits at
a table, and around him, arranged in the form of a semi-
circle, are fifteen boxes numbered 1-15, but otherwise
identical to all appearances. They differ, however, in weight,
the first box weighing 20 grams, the second 40 grams, the
third 60 grams, the next 80 grams, and all the remainder
weighing 100 grams. The subject is instructed to pick up the
first box, then the second box; his task is to say whether the
second is heavier than the first. He replies in the affirmative
and is then instructed to pick up box 2 and box number 3.
Again, the second box is the heavier, and so he continues,
always finding the second box heavier until he goes on to
boxes 5 and 6, which, of course, are both objectively equal.
Many subjects accept the suggestion implied in the
procedure that the second box is always heavier than
the first, and go on saying 'heavier' through the rest of the
trial, in spite of the objective quality of the weight of the
boxes.

(6) Memory Suggestion Test. In this test the subject is shown
a picture containing a large number of details. He is asked
to memorize the details of the picture for 15 seconds and
then the picture is taken away and he is asked a series of
questions about the contents of the picture. One of these
might be, ' Was the cat lying on the chair in the right corner
or the left corner of the room? ' There was no cat at all in the
picture, but many people accept the implication of the ques-

Can Personality be Measured? 215

tion and firmly convince themselves that there was a cat on
one of the two chairs.

These are only some of the many tests which have been
used for the measurement of suggestibility. When a sufficient
number of them are applied to groups of people and we in-
quire again, as we did in the case of persistence tests, whether
a person who is suggestible in one test is also suggestible on
the others, we find that, in this case, the answer is in the
negative. The first three tests mentioned, i.e. the Chevreul
Pendulum test, the Body Sway test, and the Arm Levitation
test, do, indeed, correlate together quite highly. The next
three tests, i.e. the Smell Suggestion test, the Progressive
Weights test, and the Memory Suggestion test also correlate
together, although the relationships are very much lower
than in the case of the first three tests. However, when the
first three tests are correlated with the last three tests, it is
found that there is no relationship whatsoever. In other
words, a person who is 'suggestible' as defined by his per-
formance on the first three tests may, or may not, be 'sug-
gestible' as defined by the last three tests; it is impossible to
make any kind of prediction because there is no relationship
at all between these two meanings of the term. We are deal-
ing with two quite unrelated types of suggestibility - prim-
ary suggestibility and secondary suggestibility, as they have
been called.

Not only are these two types of suggestibility not related
to each other; they behave quite differently in a number of
ways. Thus, primary suggestibility is not at all correlated
with intelligence; intelligent people are no less suggestible
than dull ones. Secondary suggestibility, however, does cor-
relate with intelligence. Here the more intelligent ones are
less suggestible and the dull ones are more suggestible.
Again, primary suggestibility is closely related to hypnosis;
a person who is highly suggestible also tends to be easily
hypnotizable. This is not true of secondary suggestibility. A
person who is suggestible along the lines defined by this set
of tests is neither more easily nor less hypnotizable than
a person who is non-suggestible. These are only some

2 1 6 Part Two • Personality and Social Life

of the differences observed between these two types of
suggestibility, but there are many more, and they leave no
doubt at all that we are dealing here with quite different
traits.

Nor are these the only two types of suggestibility. A third
variety, which does not seem to correlate with either of the
other two, and which has been called tertiary suggestibility,
is demonstrated in the following type of test. A social atti-
tude inventory is given to a group of people, calling for their
opinions on a variety of social issues. The filled-in question-
naires are then collected and scored. A new set of question-
naires is now constructed containing many more questions
than the old one, but incorporating all the old questions
also. This time, however, the students are told how each
question has been answered by some prestigeful person or
group, such as the President of the United States, members
of the local Chamber of Commerce, or even their own Pro-
fessor. These new questionnaires are now handed out after
an interval of about three or four months, care having been
taken to construct the hypothetical ' prestigeful ' answers for
each person in such a way that in half the cases they agree
with his previous replies, and in half the cases they contra-
dict his previous replies. In this way it becomes possible to
measure the amount of change produced in his replies by
the presentation of the answers of the prestige group or per-
son. If he is suggestible, then we would expect him to change
a number of his answers in the direction of the replies given
by the prestige group. This does, indeed, happen and people
differ considerably in the amount of suggestibility shown in
a test like this.

Thus, we find that there are at least three types of sugges-
tibility, and probably several more different types quite un-
related to those mentioned so far will be unearthed by future
experimentation. It will now be clear why, in our rating ex-
periment, there was so little agreement among different
raters of the 'suggestibility' of the subjects whom they rated.
There simply is no one trait of suggestibility which can be
rated in this way, and unless we make it clear which kind of

Can Personality be Measured? 217

suggestibility we are concerned with, we are caught in the
snares of semantic confusion. Before such semantic confusion
can be cleared up, however, we must clarify the hypothesis
embodied in the trait names by the sort of experimental
work just described. Because it enables us to do this on an
objective basis, the objective test appears to be so much
superior to either the rating or the questionnaire.

The methods discussed so far illustrate only one of the
many ways in which objective tests can be used for the mea-
surement of personality traits. It will have been noticed that
what is done is essentially this. We start out with a popular
notion of some kind of trait; we then go on to refine this
notion, express it in terms of objective tests, study its im-
plications statistically, and finally emerge with a much more
refined and accurate notion than the one we started out
with. In this respect what we are doing in the field of char-
acter and temperament is essentially similar to what had
been done previously in the field of ability. The methods of
isolating and measuring persistence and suggestibility are
precisely analogous to the methods used for isolating and
measuring verbal ability, perceptual ability, memory, or in-
telligence. The reader familiar with my discussion of these
topics in Uses and Abuses of Psychology will see the parallel
without any difficulty.

However, objective tests enable us to do considerably
more than this. In what we have said so far we have been
concerned simply with an improvement in the description
of behaviour. There has been no mention of any funda-
mental laws governing behaviour, nor of any deduction
from such laws to indicate why a person showed much or
little persistence, or suggestibility, or whatever trait we
might be concerned with. Descriptively we have succeeded
in isolating certain continua on which we can assess a given
subject's position with a certain degree of accuracy. How-
ever, this clearly is not enough, although it is a useful
first step. After all, an individual cannot do anything with
these continua except sit on them, and we would like to be
able to do a good deal more. We would like to know why a

2 1 8 Part Two • Personality and Social Life

given individual finds himself in the position he occupies ; we
would like to know how we can change his position ; and we
would like to be able to deduce all this from some general
kind of law. It will be the task of the next chapter to indicate
how this can be done and how we can arrive at more funda-
mental types of measurement. The topic is too important
and too complex to be squeezed into a few paragraphs at the
tail-end of this chapter.

Nor will we deal with another type of measurement which
is also of considerable importance. This is the field of
psycho-physiology, or psychosomatic relationships, and great
strides have been made in it in recent years owing to the
advances in electronic apparatus which can be used for the
refined measurement of slight physiological changes which
are indicative of emotion. Again, the field is too complex to
be compressed into a few sentences, and I have tried to deal
with some aspects of it in an earlier chapter, where I dis-
cussed the lie detector apparatus, essentially based on these
psycho-physiological relationships.

There does, however, remain to be dealt with in this chap-
ter a group of tests which have blossomed during the last
twenty years to a surprising extent, and which have been so
widely used that no chapter on the measurement of person-
ality could be complete without at least some mention of
them. I am referring to the so-called 'projective techniques',
which differ from the methods so far considered by taking a
conscious pride in being subjective rather than objective,
and which try to deal not with certain traits in isolation but
with what is often called the 'total personality'. The typical
result of an examination by means of one of the projective
techniques is not a rating of a given trait or a given set of
traits, but a personality description in which an attempt is
made to convey a total impression of the person who has
been tested and examined.

The actual term 'projection', as used in connexion with
these tests, is something of a misnomer. Originally, the term
was used by Freud to characterize the tendency to ascribe
to the external world repressed mental processes which were

Can Personality be Measured? 219

not consciously recognized ; as a result of this repression the
content of these processes was supposed to be experienced as
part of the outer world. We have already encountered an
example of this process of projection in our description of the
experiment on ratings in which people, unconscious of the
fact that they themselves were stingy, 'projected' this stingi-
ness on to others. As Freud would have put it, having re-
pressed any knowledge of this reprehensible trait in them-
selves, they tended to project it outwards and found evidence
for it in other people, contrary to objective fact.

While this was the original meaning of projection, the
term was broadened considerably by later writers and now
means simply a tendency on the part of an individual to ex-
press his thoughts, feelings, and emotions, whether conscious
or unconscious, in structuring some relatively unstructured
material. It is in this that the projective tests differ most
from the objective tests considered hitherto. In an objective
test there is a correct answer, a right and wrong way of doing
things, or at least a numerical measure of success and failure.
In the projective test all this vanishes. The subject may be
shown a picture and asked to write a story about the con-
tents; what situation is depicted in the picture; how the
situation came about ; how it will end ; what will happen to
the main characters, and so on. The hypothesis underlying
this method is that in telling this story the subject will in-
evitably draw on his own hopes and fears, his own conscious
and unconscious emotional complexes, and will reveal them
in some way in the story. Clearly, reading back from the
story to this hypothetical complex, feelings, and so on, is a
very hazardous matter, and equally clearly, there is no right
or wrong way of telling a story. This technique, called the
Thematic Apperception test, has been used very widely, and
there is some evidence that certain types of mental content
at least are indeed recognizably expressed in stories told.
Thus, in one experiment groups of students whose political
sympathies were known, were shown photographs depicting
clashes between police and strikers. The stories they told
about these pictures were then scored according to the

220 Part Two ■ Personality and Social Life

amount of sympathy shown with the police or the strikers
respectively, and these scores were found to be highly cor-
related with the known political attitudes of the students. In
other words, right wing students made up stories in which
the strikers had committed various crimes and were un-
reasonably resisting the loyal and patriotic police forces.
Students with opinions tending towards the left, on the other
hand, made up stories in which the brutal and licentious
police were beating down hard-working, honest men rebel-
ling against intolerable injustices.

Another test which has become very widely known is the
so-called Rorschach test. In this, a set of ink-blots, some
coloured, some just black and white, is shown to the subject,
who is given the following instructions. c People see all sorts
of things in these ink blots ; now tell me what you see, what
it might be for you, what it makes you think of.' Interpreta-
tion of the results is attempted by noting four things. First,
where on the card does the subject see whatever he claims to
see? Does he make use of the whole card, or does he just see
a tiny detail somewhere in the corner? This is called ana-
lysis by location. Secondly, the examiner tries to find the
so-called determinants of the ink blot which the subject has
used in constructing a response, such as the form, the colour,
the shading, and so forth. Thirdly, the examiner studies in
detail the content of the response, i.e. what kind of thing did
the subject see. Lastly, the popularity or originality of the
response is taken into account; some people give only stereo-
typed replies, others see highly original and unusual things
in the ink-blot.

Certain relationships have been worked out between these
various categories and personality characteristics. Thus, a
tendency to give responses based on the whole card rather
than on parts of it is supposed to indicate a tendency
towards making broad surveys of presented material, a
tendency which, if exaggerated, indicates a person fond of
expansive generalities and neglectful of obvious detail. Con-
versely, great attention to small details in the blot is sup-
posed to denote habitual attention to the concrete and a more

Can Personality be Measured? 221

practical approach. Taken to extremes it is supposed to in-
dicate pedantry and obsessions of thoroughness and
cautiousness.

The tendency for responses to be dominated by colour is
supposed to indicate habitual impulsiveness, eccentricity,
capacity for intense emotional experiences, and in extreme
cases violence and flightiness. Attention to form rather than
to colour is supposed to indicate intellectual steadiness or
introversion. Determination of responses by the shading
characteristics of the blot is supposed to indicate a consider-
able degree of repression.

A whole mythology has grown up around the Rorschach
test to such an extent that there are acknowledged 'Rors-
chach experts' in the United States who do not use any other
methods, but simply study the responses given by patients to
these ink-blots. Many psychiatrists, despairing of ever en-
compassing the complexities of neurotic behaviour in the
interview situation, have grasped at the Rorschach as the
proverbial man grasps at a straw, and pay considerable at-
tention to its verdict. Unfortunately, the evidence regarding
its validity indicates that if the various possible diagnoses
were written on the different faces of a die, and the die then
cast at random, the diagnosis arrived at by this somewhat
chancy method would not be considerably inferior to that
arrived at by the Rorschach expert in all his panoply. A
single experiment must suffice to illustrate the truth of this
observation.

In this experiment, a whole group of projection tests, in-
cluding the Rorschach, was given to prospective pilots in the
United States Air Force. These were followed up over a
period of years, and finally two groups were selected, those
who had unmistakably broken down with neurotic dis-
orders of one kind or another, and those who had made a
spectacularly good adjustment in spite of considerable stress.
In other words, out of a very large group of people, two
groups were chosen representing, respectively, those making
the best and those making the worst kind of adjustment.
Their projective test records were then taken out of their

222 Part Two • Personality and Social Life

files and given to recognized experts in the field, the instruc-
tions being to say which records would predict good adjust-
ment and which would predict poor adjustment. The ex-
perts were familiar with the criterion used, had had experi-
ence with the type of task on which they were engaged and,
on the whole, regarded it as a reasonable experiment in
which they could expect to be successful. In actual fact, not
one of the experts succeeded in predicting with better than
chance success the future performance of these airmen. They
failed when using a single test; they failed when using all the
tests together; and they failed when their predictions were
combined in all possible ways. Only one single result was
statistically significant, and that was significant in the wrong
direction !

If this result is typical of experimental attempts to validate
the Rorschach, then the reader may rightly ask why it has
achieved such universal fame and is so widely used. The an-
swer is a rather complex one, but it may perhaps be pre-
mised by a little story which will illustrate some of the mech-
anisms at work. Quite a few years ago I was studying French
literature and history at the University of Dijon. Also attend-
ing courses there was a young Viennese girl, without whose
presence at the University my French would now be much
better than it is. Unfortunately, the ratio of men to women
at this University was deplorably high and a good many
French students were offering to teach this young lady
French outside the official lecture hours. (Nobody offered to
teach me French!) However, she was very interested in
dancing, and I was fortunate enough to be allowed to take
her to the official mid-term dance, the great social event of
the town.

On the way to the dance, to my mortification I sprained
an ankle and could only hobble to the place where the fes-
tivities were being held. I desperately searched my brain for
a way of retaining the young lady's interest without having
to go on to the dance-floor, and finally hit on what still
seems to me a very ingenious idea. We were sitting down at
a large table with a rather international group of students,

Can Personality be Measured? 223

and I casually mentioned that I was an expert graphologist
and could tell anybody's character from their handwriting.
(Of course, I knew nothing whatsoever about graphology,
but a desperate situation calls for desperate remedies.) The
response was literally overwhelming. Everybody pulled let-
ters and other documents out of their pockets, demanding
that I should tell the character of the people concerned.
Many surreptitiously slipped away to write something down
themselves in order that they might at long last learn some-
thing about their own character.

To cut a long story short, I managed to retain my com-
panion's interest throughout the evening, and I probably
did more good for the renown of graphology as an exact
science than I have ever been able to do damage to it in my
more professional writings. About 95 per cent of the c clients'
reported themselves amazed at the uncanny accuracy of my
characterizations. I think any scientist who would have
come in to challenge the claims of graphology at that time
would have had a very hard time indeed. What happened?

When we write a personality characterization there are a
number of factors which are acceptable to the person to
whom it is meant to apply, although objectively there may
be no connexion at all. In the first place, there are a number
of traits which most people think they possess, although in
reality they may not possess them at all. An example of this
has already been given when it was pointed out that 98 per
cent of the population consider that they have an above
average sense of humour. If, therefore, we want to write a
description which will be acceptable to almost anybody, we
would merely have to introduce a sentence like ' You have a
very good sense of humour'; 98 per cent of the population,
at least, will agree with us that this correctly describes them
and will marvel at the accuracy with which we have been
able to diagnose their handwriting, or read their Rorschach,
or analyse their Thematic Apperception test. Similarly,
most people at times have feelings of insecurity ; most people
feel that their real worth is not always being appreciated and
that they have sometimes been pipped at the post by people

224 Part Two • Personality and Social Life

less able than they. Just fill your whole personality descrip-
tion with universally acceptable statements of this kind, and
everyone will recognize his own picture in them.

That this is true has been experimentally demonstrated
on several occasions. The experimenter gives his students an
outline, say, of the beliefs of graphologists or of Rorschach
experts; he then asks them to submit a sample of their hand-
writing, or actually to undergo the Rorschach test. He takes
the records away with him and after a few days hands out to
each member of the class a typed statement of what the
Rorschach or the graphology specimen has revealed about
each student's character. The students are given a few min-
utes to read through their characterizations and are then
asked whether they consider these to be accurate descrip-
tions of their own personalities. Usually ninety to ninety-
five out of one hundred raise their hands. The experimenter
then asks one of them to read out his own characterization.
All the others then realize that each one of them has been
given the same personality description and that what they
all agreed to as being representative of themselves was, in
fact, an overall set of traits applicable to practically every-
one.

The second factor working in favour of the analysis is the
vagueness and ambiguity of the terms used. The persons
whose characters are being analysed almost inevitably pick
out that meaning of a term or phrase which they consider
applicable to themselves, forgetting all the other meanings
which might not be so applicable. These two factors work
particularly strongly in the case of neurotic and psychotic
patients, where the Rorschach is most frequently applied. It
is perfectly safe to say in every case that the patient is
anxious or depressed ; if he is not overtly so, then it can al-
ways be argued that some other symptom acts as a defence
against his anxiety which thus remains unconscious. This
policy of 'heads I win, tails you lose', which is so character-
istic of psychoanalysis as a whole, has been triumphantly ap-
plied by the Rorschach experts, and serves to make any ex-
perimental examination of their tenets difficult.

Can Personality be Measured? 225

We must therefore rule out entirely the personality de-
scription as being in any way an acceptable proof of the
accuracy of the projective type of technique. We can see
why graphologists, palmists, and other self-styled scientists
are so successful in bamboozling the public, and why astro-
logers can still persuade the more gullible members of the
public of their occult powers. The fact that clinical psycho-
logists and psychiatrists have fallen for a similar type of trick
does not argue too well for their critical acumen and
scientific outlook.

If we rule out the personality characterization, what then
are we left with? Frequently another method is used which
has become known as the matching method. In this an at-
tempt is made to rule out the many sources of error inherent
in a simple assessment of personality description. What is
done instead is this. Five patients, say, are given the Ror-
schach test; their records are then analysed and the person-
ality descriptions resulting therefrom are handed over to an-
other expert who is also given the case records of the same
five people. His task, then, is to match the case records and
the personality descriptions. If he succeeds in doing so, then
it is argued there must be some truth in the personality de-
scription, because otherwise how could a correct matching
be obtained?

Unfortunately, this method also is subject to so many
sources of error as to be practically useless. It is very fre-
quently possible to get from the record indices of a person's
background, intelligence, and upbringing which may serve
to identify him, but which are completely independent of
the purpose of the test. Thus, in one experiment dealing
with graphology, which I carried out myself, the subject was
asked to copy the questions of a questionnaire which he also
had to answer ; the answers were then cut off and the hand-
writing specimen supplied to an expert. One subject in num-
bering the questions left out number 13 and put 12a instead.
No wonder he was called 'superstitious' by the expert and
recognized and correctly matched because of this single
adjective.

226 Part Two • Personality and Social Life

Another person in a similar study of the Rorschach gave
many anatomical responses and was correctly identified as a
medical student, the only one in the sample of five.

What is required in studies of this kind is to have the
matching done, not only by the expert in graphology, or the
Rorschach, or the Thematic Apperception test, but also by
a very intelligent person who is quite ignorant of the rule of
interpretation of these various tests. I have found in a num-
ber of investigations that by relying entirely on external cues
of the kind mentioned, such an independent observer was
actually more successful in performing correct matchings
than Rorschach and graphology experts! Without control
experiments of this kind matching is not a safe method to use
as evidence.

What has been said about the Rorschach applies in very
much the same way to the other projective techniques which
have been used at various times. As the reader may be inter-
ested in learning something about the kinds of methods
which have been used, a few of them will be enumerated.
An auditory version of the Thematic Apperception test is
the so-called Tautophone test, or 'verbal summator5. This
consists of a set of phonograph records on which a series of
vowel patterns are repeated. These records are played to the
subject at a low intensity and at some distance, so that an
illusion is built up that the subject is actually listening to
human speech. Instructions are to report what the man on
the record is saying as soon as it can be understood. The sub-
ject's record is then analysed in some detail.

Another method, rather similar to the Thematic Apper-
ception test, is that of story-telling, in which subjects are
asked to tell stories about people or situations which are de-
scribed to them, or even about musical pieces or odours
which are used to stimulate fantasy. Alternatively, a story
may be begun and the subject left to finish it.

A slightly more analytical type of test is the incomplete
sentences test. The beginnings of some sentences are given
to the subject, and he is then required to write in the end-
ings. Such beginnings as the following are used :

Can Personality be Measured? 227

My hero is

I worry over

I become disgusted with

I feel ashamed when

I fail

My father used to

The men around here ..*

It is hoped in these cases that the continuation written in by
the subject will reveal something about his attitudes, fears,
and complexes.

Handwriting analysis is another one of the so-called pro-
jective techniques in which various characteristics of a per-
son's handwriting are used to arrive at a personality de-
scription of him. Thus, for instance, anxiety is supposed to
be shown by narrow distances between words; narrow dis-
tances between lines; lines beginning at the extreme left
without a margin; slants and flourishes to the left; heavy
pressure or irregular losses of pressure; and small slow writ-
ing with occasional abrupt losses of height. While there is
some slight evidence that graphology may be worthy of
further study, at the present time it can certainly not be con-
sidered to be either a reliable or a valid technique.

Drawing, painting, and play have all been used as pro-
jective techniques. Interpretations are made of the drawings
or paintings the child makes, either quite spontaneously or
of certain suggested subjects, such as a tree, or a house, or a
man. Similarly, play may be analysed in a standard situa-
tion in which the subject is supplied with a number of dolls
representing the father, the mother, the baby, and so on.
When the child pushes the father doll's head into the toilet,
for instance, this may be interpreted as denoting aggressive
feelings towards the father!

What strikes one most about all these projective tech-
niques, apart from the fact that little attempt is usually made
to show that they are valid indicators of personality, and the
further fact that negative results are quietly disregarded by
the 'experts' making use of these techniques, is the fact that
they seem to be based on a logical fallacy. The argument in

228 Part Two • Personality and Social Life

their favour is usually put something like this. Everything
we do, it is said, is determined by conscious or unconscious
mental processes of one kind or another. According to this
basic 'dynamic' theory, anything that we do may be used to
argue back to the causal factors which are responsible for
our action. If that be so, then the best techniques for diag-
nosing personality would clearly be those which will give the
greatest freedom of choice in structuring our environment.
It is for this reason that stories, paintings, play activities,
Rorschach interpretation, and so on, are considered so
valuable.

This argument appeals to many people because the major
premise of it is probably quite true. There is, of course, no
clearly convincing proof of this, but it is not unreasonable to
suppose that most of our major activities are, in fact, per-
formed for some reason, or that this reason is in some sense
diagnostic of our personality. Because most of the activities
studied by the expert in projective techniques are caused by,
and diagnostic of, features of personality, it does not follow
that we are able to argue back from the finished product to
the causal mechanism. Perhaps I can best explain this by
means of a somewhat fanciful example.

It is often said that Jaguar sports cars are bought by sport-
ing young men. In other words, buying a Jaguar sports car
is regarded as a kind of projection test in which the personal-
ity of our * sporting young man' finds an outlet for its various
characteristic features in the speed, power, etc., of this par-
ticular car. Let us assume that this argument is correct.
What the projective technique expert now asserts is that we
can argue back from the purchase of a Jaguar sports car to
the possession of these various psychological features. This,
however, would logically be true only if we could say that
only sporting young men buy Jaguar cars. However, it is pos-
sible, and indeed probable, that many other types of people,
for many other types of reasons, also buy Jaguar sports cars,
thus completely invalidating the inversion of the original
argument. Thinking of the relatively limited number of
people in my acquaintance who have bought Jaguar sports

Can Personality be Measured? 229

cars, I found the following reasons for the purchase : the first
one was an American business man who would have pre-
ferred a more orthodox family car, but who said that the re-
sale value of a sports car in the United States was very much
higher than that of a British family car. Another business
man bought it because, as a salesman, he had to get around
the country very quickly and make long trips over to the
Continent, where he found the speed invaluable. A third
one bought it to impress his girl friend. Presumably, there
are many more reasons why a person might buy a particular
car, reasons which would make it extremely hazardous to
argue back from the purchase to his personality.

But it is precisely this logically fallacious argument which
underlies all the projective techniques. Because, it is said, a
person of vehement emotion makes use of the colour of the
Rorschach in his choice of interpretations, therefore a per-
son who makes choice of colour in his interpretations must
have strong emotions. Even if the first part of the argument
were true, which it is not, the second part would by no
means follow. There are many other reasons which might
cause a person to be particularly conscious of the colour of
the blot, and which might lead to quite different views of
the subject's personality if they were taken into account.

In brief, then, the projective technique experts very much
call to mind the famous story of Tartarin de Tarascon.
Those familiar with Daudet's classic will remember that
Tartarin was the greatest hunter in that lovely little town in
the Midi. Unfortunately, as there were no animals to hunt in
the neighbourhood, all that the Tarasconais could do when
they sallied out on a bright sunny morning was to throw
their caps into the air and shoot at them ; the one who had
the most holes in his cap was the winner. Tartarin, not con-
tent with the reputation he gained in this way, wanted to ex-
tend his field of operations and go lion-hunting in Africa.
The more he talked about this, and the more vividly he de-
scribed the exploits he would be going to accomplish, the
more he became convinced that he had actually been there
and was describing feats actually accomplished. For a long

230 Part Two • Personality and Social Life

time his fellow townsmen fell under the same spell, but
finally it was broken and he had to carry out his great plans
in actual fact.

His adventures in Africa were many and various; the
most important of them, perhaps, was that he did finally
manage to shoot a blind, old, and mangy lion who had been
used by the Arabs very much as a dog is used by beggars to
hold a cap before passers-by to throw their coins into. This
miserable animal was skinned and the hide sent back to
Tarascon. When he finally arrived back, followed by a
camel which he had somehow managed to acquire, he was
rapturously greeted by the townsfolk, who had blown-up the
fact of this one, mangy, moth-eaten skin into a glorious vic-
tory over hundreds of wily and dangerous kings of the
desert.

In very much the same way, projective techniques ex-
perts started out by emphasizing the ways in which they
would provide validation for their methods. Gradually they
persuaded themselves, and those whom they could induce
to listen to them, that the validation had already been ac-
complished. However, when finally the realization dawned
that nothing of the kind had really happened and they were
forced to try to hunt for some tangible and palpable evi-
dence, they set out boldly, only to return with mangy,
moth-eaten, and useless data, based on logical fallacies and
completely lacking even the most elementary types of con-
trol necessary in this kind of work.

6

PERSONALITY AND CONDITIONING

This chapter deals with a series of experiments and theories
due to a man who was perhaps the greatest psychologist in
the brief history of that science. Pavlov himself, curiously
enough, considered himself to be a physiologist and had an
exceedingly low opinion of psychology altogether. Neverthe-
less, his great contribution, by general consent, has been to
psychology and his physiological speculations have been re-
ceived very coldly indeed by his fellow physiologists.

Most people have heard something of his work and the
term 'conditioning', originally introduced and defined by
him, has become fairly widely known, yet the true import-
ance of his work is seldom realized, and even among psycho-
logists many misconceptions are rife. A good rousing sum-
mary and elaboration of these misconceptions can be found in
the writings of George Bernard Shaw, particularly his chap-
ter on 'The Man of Science' in Everybody's Political What's
What, and in The Black Girl in Search of God.

What Shaw has to say is interesting, although in fact it is
almost complete nonsense. It is interesting because it shows
clearly how a highly intelligent person can utterly mis-
understand the aim and purpose of scientific experiments.
This lack of understanding of scientific methodology and
purpose will be startling to anyone with even a modest back-
ground in scientific discipline, but it is worth taking seriously
because it is so widespread.

What then is Shaw's argument? To him Pavlov is 'the
prince of pseudo-scientific simpletons'. All he did was 'to
devote twenty-five years of his life to experiments on dogs, to
find out in the data for his biological theory, whether their
mouths watered, and if so, how much (counted in drops of
saliva) when they had certain sensations, such as the sight or
smell of food, the hearing of words or noises, the feeling of
certain touches or the sight of certain persons or objects'. In

231

232 Part Two • Personality and Social Life

this way he discovered that habits or 'conditioned reflexes'
could be produced by association. 'This remarkable dis-
covery', so Shaw makes Pavlov say, 'cost me twenty-five
years of devoted research, during which I cut out the brains
of innumerable dogs and observed their spittle by making
holes in their cheeks for them to salivate, instead of through
their tongues.' 'Why did you not ask me?' said the black
girl. ' I could have told you in twenty-five seconds without
hurting those poor dogs.' 'Your ignorance and presumption
are unspeakable', Shaw makes Pavlov reply. 'The fact was
known, of course, to every child, but it had never been
proved experimentally in the laboratory, and therefore it
was not scientifically known at all. It reached me as an un-
skilled conjecture; I handed it on as science.'*

We may discard in Shaw's argument his emotional ob-
session with Pavlov's alleged cruelty to his dogs and also the
somewhat patronizing attitude Shaw takes to Pavlov's work,
implying that if only he, Shaw, could have taken off a few
weeks from his important work to set Pavlov on the right
track, then everything might have been all right. Shorn of
these red herrings and of the rhetorical bombast which Shaw
uses to twist the argument, he is essentially saying this : Pav-

* That some of the facts of conditioning were known previous to Pav-
lov's work cannot, of course, be doubted. As an example we may perhaps
quote from the play by the famous Spanish playwright, Lope de Vega,
which was written about 1615 called El Capellan de la Virgen. This is the
story, somewhat freely translated : ' Saint Ildefonso used to scold me and
punish me lots of times. He would make me sit on the bare floor and eat
with the cats of the monastery. These cats were such brutes that they
took advantage of my penitence. They drove me mad stealing my
choicest morsels. It did no good to chase them away. But I found a way
of coping with the beasts in order to enjoy my meals when I was being
punished. I put them all in a sack, and on a pitch black night took them
out under an arch. First I would cough and then immediately beat the
daylights out of the cats. They whined and shrieked like an infernal pipe
organ. I would pause for awhile and repeat the operation - first a cough,
and then a thrashing. I finally noticed that even without beating them,
the beasts moaned and yelped like the very devil whenever I coughed.
I then let them loose. Thereafter, whenever I had to eat off the floor, I
would cast a look around. If an animal approached my food, all I had
to do was to cough, and how that cat did run ! '

Personality and Conditioning 233

lov discovered that by occurring at the same time, two
events became linked in the mind of a person or a dog, and
that in this way habits are created. This fact is known to
everyone, but it took Pavlov twenty-five years to discover it.
What fools these scientists be !

One might in a very similar way dismiss the contribution
which Newton made to physics. The argument would run
something like this : ' Objects when left unsupported tend to
fall to the ground. This is known to everyone, but it took
Newton twenty-five years to prove it.' Most people will be
able to see the weakness in this argument; precisely the same
weakness is contained in the argument as applied to Pavlov.
Of course we all know that things fall to the ground, but few
of us know the laws according to which they do so. We all
know that habits are formed, but few of us know precisely
how they are formed and how they can be broken. The fact
that we have a vague acquaintance with certain natural
phenomena can hardly be taken to mean that a scientific
study of these phenomena is unnecessary. Common sense may
vaguely recognize the sort of thing that is happening, but
science requires more than that. It requires description and
explanation. These two terms and the meaning which they
have in science are so important that a brief discussion
appears indicated.

Fundamentally, description and explanation cannot be
considered to be entirely different processes. How do we ex-
plain why objects fall to the ground? We explain it in terms
of the law of gravitation. How do we find the law of gravita-
tion? We discover it by the detailed 'description' of things
falling to the ground. Explanation is simply the reference
back of the individual facts to general laws, laws which in
turn are derived from the detailed observation and descrip-
tion of individual facts. Some of these individual facts may
be known to common sense, and common sense may even
be able to make some vague kind of generalization. How-
ever, this is never sufficient for the scientist. The common-
sense descriptions are vague and couched in words rather
than in numbers, and do not take account of many possible

234 Pari Two • Personality and Social Life

sources of error. Common-sense generalizations are vague,
intuitive, and often contradictory. It is only necessary to
compare the common-sense generalization 'objects tend to
fall to the ground when left unsupported5 with the actual
formula giving the behaviour of falling bodies as \gt2 to
realize the tremendous difference between common-sense
concepts and explanations and scientific laws and formulae.

The main difference between description and explana-
tion, then, is essentially one of breadth and latitude; de-
scription is essentially of individual phenomena, explanation
is in terms of laws derived from large numbers of individual
phenomena and applicable to literally infinite numbers of
further individual events. It is these laws which science is
always reaching out for, and an advance in their direction is
of the greatest possible importance in the development of the
science. In arriving at such laws and generalizations it often
becomes necessary also to invent or discover certain con-
cepts which are of a peculiar abstract nature. Newton's
gravitational force was such a concept; Pavlov's 'condition-
ing' is such another. These terms do not refer to actual ob-
servable objects or events, but to hypothetical constructs
which make thinking about observable events easier and
which may enter into our equations describing and ' explain-
ing' the behaviour of objects, animals, or human beings.

This notion of explanation by reference to general laws
should be contrasted with a rather different notion which is
frequently encountered and which is particularly apposite
to psychological phenomena. We often think we have 'ex-
plained' somebody's behaviour when we succeed in showing
that it was in some ways similar to behaviour with which we
were already intuitively familiar. Thus, we all know what it
is to feel annoyed and to want to strike another person whom
we consider to have been responsible for this annoyance.
When we, therefore, find one person striking another, and
when we further learn that this other person has annoyed
the one whose aggressiveness we are trying to explain, then
we feel that we have a complete explanation of the situa-
tion. 'He was annoyed', we say, 'and therefore he knocked

Personality and Conditioning 235

down the chap who annoyed him.' Generally this is not in
any sense an explanation of what happened. People often
get annoyed without knocking down other people. Why,
then, in this particular case, did the aggressive act happen,
while in another case it did not? We might say that in the
one case the annoyance caused was greater than the other,
but if we are asked how we know this, we can only say that
it must have been so, because in the one case violence
ensued, while in the other case it did not. Reasoning along
these lines is clearly unsatisfactory. All scientific explanation
is circular, and consequently we could hardly complain in
this case that the reasoning is circular. Our objection is to
the fact that the circle is too narrow and small ; at no point
do we have general concepts and laws entering in which are
defined objectively and independently of the facts which they are
called upon to explain.

The law of gravitation is derived from the observation of
falling bodies and is in turn used to explain the behaviour of
falling bodies. So far the argument must be conceded to be
circular, but in accounting for the speed with which a given
brick - which I dropped from the roof-garden of the Mauds-
ley on to the tennis-court - fell, I do not make use of any in-
formation obtained from this particular brick. I refer back
to a general law discovered when Galileo dropped various
objects from the Leaning Tower of Pisa, three hundred years
ago. But in accounting for the aggressive behaviour of a
given person in terms of the strength of his annoyance, I
make use of his actual behaviour to indicate his degree of
annoyance. There is no general law here, but simply a beg-
ging of the question. An actual example of how we can go
from the popular man-in-the-street kind of psychological
explanation, often called a 'mentalistic' explanation by
Pavlov and his followers, to the more fundamental and
scientific explanation which has become possible through the
work of the Russian School, may be useful as an illustration.

You are at the dentist's, and the drill is biting into one of
your molars, suddenly touching the nerve; you grip the
arms of the chair as strongly as you can and you may notice

236 Part Two • Personality and Social Life

a slight lessening in the pain experienced. This, at least, is
the reaction of some people. Others may seek distraction by
pinching their thighs, by digging their finger-nails into their
hands, or even by doing complex problems in spherical trigo-
nometry in their head. Again the common-sense explanation
as always is readily forthcoming. Our attention, so it is said,
is distracted from the pain, and consequently we feel it less
sharply. The greater the distraction, the greater will be the
lessening of the pain. How do we know which distraction is
greater and which is less? Well, simply by noticing the de-
gree to which the pain is lessened. This is simply another
example of circular reasoning in which any general term is
conspicuously missing. The lessening of the pain is attri-
buted to the distracting stimulus, and the strength of the dis-
tracting stimulus is measured by the lessening in pain. There
is no possibility of proving or disproving such a theory be-
cause no independent verification is possible. What, then,
can we do to get away from this mentalistic interpretation
and achieve a kind of theoretical formulation which is cap-
able of generating broad generalizations and which can be
submitted to the ordinary scientific processes of proof?

We may start with a perfectly general law discovered by
Pavlov and called by him 'Negative Induction'. (Actually
to be strictly accurate, the first mention of such a law is due
to the German psychologist and physiologist Hering. It did
not, however, play any considerable part in his theory and
he did not use it to anything like the same systematic extent
as did Pavlov.) By Negative Induction, Pavlov meant the
experimental demonstrable fact that a positive stimulus ap-
plied in one part of the brain may cause the depression of
activity in other parts of the brain. The kind of defining ex-
periment which he uses to demonstrate this is as follows : A
dog has been conditioned to respond with a flow of saliva to
a metronome giving 120 beats a minute. By saying the dog
has been 'conditioned' is meant just this: the dog, standing
on a table in a sound-proof room, strapped in a harness
which makes it impossible for him to jump off the table, can
be presented with various stimuli by the experimenter, who

Personality and Conditioning 237

is standing outside the room. He can observe the animal
through a one-way screen, without being himself observed,
and he can count, through an electrical transmission system,
the number of drops of saliva secreted by the dog. When the
experimenter sets the metronome going, the dog can be seen
to glance towards the source of the sound, but does not saliv-
ate. The metronome, therefore, is what may be called a
neutral stimulus, i.e. it does not produce the reflex which is
being measured. After a few repetitions of the stimulus, the
dog ceases even to look towards it and becomes quite inured
to the presence of the metronome. Now a piece of food is
presented to the dog, which has not been fed for several
hours. Considerable salivation takes place. Food, then, is the
unconditioned stimulus, i.e. the stimulus which without
training brings forth the response or reflex. (The proper
translation of Pavlov's terminology would be 'conditional'
and 'unconditional' rather than 'conditioned' and 'un-
conditioned'. What we call a conditioned response is one
which is conditional upon training; what we call an uncon-
ditioned response is one not conditional upon training. Simi-
larly, an unconditioned stimulus is one which produces its
effect without any intervening training period.)

Now the conditioning procedure commences. A number
of training trials are given, in each of which the sound of the
metronome is followed by feeding the dog a relatively slight
amount of food. After ten or so training trials, the metro-
nome is presented without any food, and now it is found to
have acquired some of the properties of the unconditioned
stimulus because now salivation follows the metronome
alone. Further training trials increase the strength of the
association until finally salivation to the metronome alone
produces as many drops of saliva as does presentation of the
food. The previously neutral stimulus, i.e. the metronome,
by being paired with the unconditioned stimulus, the food,
has itself become conditioned and produces the salivary
reflex.

We are now in a position to follow Pavlov's experiment
for Negative Induction. The dog having been conditioned to

238 Part Two • Personality and Social Life

the metronome, another neutral stimulus (a buzzer, say) is
now presented to it and reinforced by food each time it is
presented. After a number of presentations, four drops of
saliva are produced when the buzzer is applied. Now, for
just one trial, the metronome stimulus is presented to the
dog (together with food) and then the buzzer. This time no
salivation whatsoever follows the buzzer. In other words, the
presentation of a strong positive conditioned stimulus has in-
hibited the occurrence of the rather weaker conditioned re-
sponse associated with the other stimulus. This, then, is an
example of Negative Induction, or the inhibition of action
in one part of the brain caused by action elsewhere. Let us
now apply this conception to the dentist's chair. A strong
stimulus (gripping the arms of the chair, pinching oneself, or
thinking of some difficult problem) is applied by the victim
at the same time as the drilling proceeds. By the law of Nega-
tive Induction we would expect this new stimulus to produce
inhibition in the remainder of the brain and thus to lessen
the pain of the dentist's drill.

The reader may object to this explanation as being purely
anecdotal and presenting nothing but an analogy. Such an
objection is perfectly reasonable and correct, but it does
leave out one vital feature. We are now in a position to take
our problem into the laboratory, to quantify the variables
concerned, and to conduct an experimental test of the ade-
quacy of our explanation. What we have stated is merely a
hypothesis; we can now go on to prove the correctness of
various deductions which can be made from this hypothesis.

First of all we must quantify the variables concerned. It is
difficult to measure the pain caused by a dentist's drill -
which in any case is constantly variable; it is also difficult to
measure the strength of the grip exerted on the chair by the
hypothetical subject of our illustration. Consequently, let us
substitute for the dentist's drill a psychological torture instru-
ment, the dolorimeter, which measures pain in units called
dols. This consists essentially of a beam of light concentrated
through a system of lenses on the forehead of the subject of
the experiment. By increasing the amount of light, the ex-

Personality and Conditioning 239

perimenter also increases the amount of heat generated, and
the illuminated spot on the subject's forehead will begin to
feel warm, and finally warmth will give way to the sensation
of pain. This pain can then be increased until the experi-
menter's research is completed, or until the subject decides
that psychological research is not his cup of tea ! The amount
of heat produced on the subject's forehead by this instru-
ment can be measured with great exactitude, as it is propor-
tional to the current fed into the apparatus. So much, then,
for the quantification of the pain stimulus.

The grip on the chair is easily replaced by asking the sub-
ject to grip a dynamometer. A description of this apparatus
has already been given in a previous chapter; it simply
consists of a steel spring which has been compressed by the
subject; attached to this spring is a dial with a pointer indi-
cating the amount of pressure exerted. This quantifies the
inhibiting stimulus.

Lastly, we must somehow quantify and objectify the very
subjective notion of pain. As long as we have to rely on our
subject's saying 'Oh! Ouch!' or something similar when-
ever the stimulus begins to be painful, it can always be ob-
jected that perhaps he was not paying attention sufficiently,
or that other subjective sources of error determined his re-
action. Fortunately, there are several objective indices
which indicate the exact moment when pain is experienced.
One of these is the psycho-galvanic reflex, which is discussed
at some length in a previous chapter. More useful still, in this
connexion, is the so-called pupillary reflex. When pain is
being inflicted on a person, his pupils get larger reflexly,
and it has been shown that in the dolorimeter a slight in-
crease in the size of the pupil can be demonstrated just before
the subject reports the first sign of pain. In other words, we
have here an objective indicator of the pain threshold of an
individual at any time.

We are now ready for our experiment. The subject is sit-
ting at a table, his forehead blackened with burnt cork; the
dolorimeter is sending out its rays at relatively low intensity,
so that the subject's sensations are of warmth rather than of

240 Part Two • Personality and Social Life

pain. At the same time a film camera focused on his eyes is
taking pictures of his pupils. Now the current in the dolori-
meter is increased until the experimenter discovers that the
pupil of the subject has increased in size. The experiment is
then stopped, the film is developed, and the exact point
determined when the subject's pupil first begins to show an
increase in size. This is then correlated with the setting of
the dolorimeter at that moment, and we thus obtain an
estimation of the pain threshold of this subject, i.e. the pre-
cise amount of heat required to give rise to the sensation of
pain. A number of readings are, of course, taken, in order to
have a more reliable estimate, but they are precisely alike in
the method of determination of the pain threshold.

The whole procedure is now repeated, with one important
difference. This time the subject is required to pull the
dynamometer at a certain rather low level of strength, say
with a pull equivalent to ten pounds. Again a number of
determinations of his pain threshold are made, the theory
being that through Negative Induction the pain threshold
should now be higher, i.e. the subject should be capable of
experiencing a slightly higher degree of heat without show-
ing the pupillary reflex indicative of pain. This is indeed
found to be so, and the experiment is repeated with increas-
ingly strong pulls on the dynamometer. A pull often pounds
is followed by one of twenty, then of thirty, then of forty, and
then of fifty pounds. Each time the theory predicts a further
rise in the pain threshold of the subject, and each time such
a rise is indeed observed. When the experiment is over we
can plot the increase in pain threshold as a function of the
amount of pull exerted on the dynamometer, or in Pavlov's
words: 'the amount of Negative Induction as a function of
the positive stimulus given'. In this way we can quantify the
variables with which we are dealing, study the functional
relationships between them, and demonstrate the adequacy
of our theoretical formulations to deal with phenomena of
this type. The advance over mentalistic explanation in terms
of concepts like 'attention' and 'distraction' is obvious.

In the explanation which we have given of the pheno-

Personality and Conditioning 241

mena of attention and distraction, the concept of Negative
Induction has been used as if it did not allow for any further
analysis. This, however, is not true. Pavlov has in fact car-
ried the matter just a step further, but before we can discuss
his theory of excitation and inhibition we must look at an-
other set of experiments which led him to postulate the
existence of these two profoundly important properties of
the cortex. This will not only give us a better understanding
of the experiments just described, but it will also lead over
directly to an explanation of certain extremely important
determinants of personality differences.

Let us go back, then, to the original conditioning experi-
ments in which the dog becomes conditioned to neutral
stimulus by pairing this with an unconditioned stimulus.
What are the conditions under which the conditioned reflex
is formed? In the first place, the animal must be in a good
state of health ; the dog that is not well, or is in active pain does
not form conditioned reflexes easily or well, possibly because
of Negative Induction. The dog must be in an alert state;
a dog that is drowsy or sleepy may not even notice the neu-
tral stimulus, and therefore will not be able to connect it up
with the food. The dog should be hungry; in a state of re-
pletion even food will not produce salivation, and conse-
quently there is little hope of conditioning the neutral stimu-
lus to do so. There should be an absence of strong disturbing
stimuli, as otherwise Negative Induction would inhibit the
formation of the conditioned reflex. The conditioned stimu-
lus itself should be neither strong nor unusual ; if it is either,
it will itself produce Negative Induction, and therefore in-
hibit the formation of the conditioned reflex.

Lastly, the external stimulus which is to become the signal
for a conditioned reflex must overlap in point of time with the
action of the unconditioned stimulus. In other words, it is no
good putting on the metronome, switching it off again, and giv-
ing the dog the food ten minutes later. Nor would it be any use
giving the dog the food and then switching on the metronome.

Actually, there is an exception to this general rule. This
exception, however, is a very important one, as it shows how

242 Part Two • Personality and Social Life

a whole chain of conditioned reflexes can be established.
Supposing we have conditioned our dog to salivate at the
sound of a metronome; we now wish to make him salivate
whenever we switch on a dim light in front of his eyes. It
would be easy to achieve this by pairing the light and the
food stimulus. It can also be done, however, without ever
establishing any contact at all between the light and the
food. The method is as follows: switch on the light, then
switch it off again, allow a few seconds to elapse, and then
put on the metronome. After a number of repetitions of this
association between light and metronome, the light itself
produces salivation. This is called the establishment of a
secondary conditioned reflex, or a reflex of the second order,
and it has proved possible to go further than this and to
establish conditioned reflexes of still higher orders. Thus, we
could now condition the dog to respond with salivation to a
buzzer by pairing the buzzer with the light in the manner
outlined above. Through the establishment of conditioned
reflexes of the higher order, the network of connexions in the
cortex can thus become exceedingly complex.

What stimulus may be used to produce a conditioned re-
flex? The list is almost endless. It could be any agent what-
soever in nature which acts on any adequate receptor or-
gan. It could be any change in any such stimulus, i.e. in-
creasing the loudness of a tone, or decreasing the brightness
of a light. It could be any combination of stimuli, e.g. the
sound of a buzzer and the switching on of a light. It could be
the cessation of a stimulus, i.e. turning off a light or a buzzer.
It could be a simple time interval; thus by feeding a dog
every half an hour, the dog will, after a while, begin to saliv-
ate shortly before he is due to be fed. Of particular interest,
however, is another type of stimulus known as the trace re-
flex, which is important because it again extends the range
over which the conditioned reflex can work. Thus, a dog can
become conditioned to the time elapsing from the applica-
tion of the stimulus to the giving of the reward, i.e. the food.
If we condition a dog to a buzzer applied at the same time
as the dog is being fed, thus establishing a conditioned reflex

Personality and Conditioning 243

to the buzzer, we can transform this into a trace reflex by
gradually increasing the interval between sounding the
buzzer and feeding the dog. Thus a trace reflex extends the
possible action of conditioning in point of time, just as a
secondary conditioned reflex extended it from the point of
view of different stimuli.

One further method of increasing the range of conditioned
reflex activity must be noted. Suppose that we condition a
dog to salivate to a tone given off by a tuning-fork vibrating
at a rate of a thousand vibrations a second; suppose we now
sound another tuning-fork, vibrating at a rate of 800 vibra-
tions a second; will the dog salivate? One might argue that
he has been conditioned to a sound, and that therefore he
should salivate. One might also argue that he has been con-
ditioned to a particular sound and that as the sound now
presented is a different one, he should not salivate. In actual
fact, the dog will not stand on principle, but will compromise
between these two extremes. He will salivate, but rather less
than he would to the exact tone. Quite generally, the greater
the similarity between the conditioned stimulus and the
one applied to the dog, the greater will be the strength of the
conditioned reflex. An actual example from Pavlov's work
will make this clear. A dog had been conditioned to salivate
when touched on the thigh. He also salivated on being
touched elsewhere on the body, but the further away from
the thigh the touch was applied, the less was the salivation
elicited (Table 2 below will show this most clearly).

table 2

hind paw 33 drops

thigh 53 drops

pelvis 45 drops

middle of trunk 39 drops

shoulder 23 drops

foreleg 21 drops

front paw 19 drops

This general rule is known as the law of stimulus general-
ization. In other words, by conditioning a particular stimu-
lus Sv we also simultaneously condition a whole group of

244 Pari Two - Personality and Social Life

stimuli 6*2, £3, SA . . . . Sn, in such a way that a conditioned re-
flex is obtained to any of them in proportion as they re-
semble the original stimulus. Without this law the whole
process of conditioning would be of purely esoteric interest;
exact identity of stimuli is difficult enough to produce in
the laboratory, and would be quite impossible to achieve in
everyday life. If, therefore, conditioning was dependent on
identical repetition of stimuli, its range of application would
be so small as to be minimal. Through stimulus generaliza-
tions, however, it becomes extended and applicable to every-
day events.

All the phenomena discussed so far are treated by Pavlov
under the general term of ' excitation5. By this he means
essentially that when a stimulus is applied to any sense organ
- a light shone into the eye, a sound striking the ear, a
touch applied to the skin - nervous excitation is produced
which travels through the central nervous system to the cor-
tex and there interacts with any other kind of nervous ex-
citation produced simultaneously. The precise neurological
details of what happens are largely unknown and in any
case do not concern us, as this is not a book on neurology.
We may note, however, that there is ample neurological
evidence that a passage of a nerve current through the chain-
work of nerves does not leave this chain- work in its former
state, but tends to modify it in certain ways. This modifica-
tion appears to take place largely at the synapses, i.e. the
points where different nerves or neurons meet, and where
the nervous impulse is handed over from one set of neurons
to another. These synapses are a kind of switching station,
and the passage of the nervous impulse of a certain strength,
and going in a certain direction, has certain lasting effects
on the synapses which make the subsequent passage of a
similar nerve impulse in the same direction easier. It is prob-
able that these semi-permanent modifications occurring at
the synapses are at the basis of the phenomena of learning,
habit and conditioning, although it would almost certainly
be wrong to say that a given habit is in any sense located in
a particular synapse or set of synapses. The picture is very

Personality and Conditioning 245

much more complicated than that, but for the purposes of
this chapter enough has been said to give the reader an idea
of what is meant by ' excitation ' in the context of Pavlov's
theory.

Pavlov's discovery of some of the facts of excitation is im-
portant enough, but far more important, novel, and fruitful
have been his discoveries of a large group of other facts
which complement those of excitation and go by the name
of inhibition. The necessity of postulating such a comple-
mentary set of factors became obvious quite early in Pav-
lov's work and was forced on him by facts which have been
reproduced since in many laboratories.

One type of inhibition we have already encountered. It is
called * external inhibition ' by Pavlov, and its consideration
was forced on him when he found that conditioned reflexes
could very easily be disrupted, or even completely inhibited,
by strong external stimuli. During the first years of his work
he did not realize the necessity of having sound-proofed
laboratories, and the many sights and sounds of the open
rooms in which he tried to condition the dogs produced so
many distracting stimuli that it was very difficult indeed to
obtain worthwhile results. External inhibition of this type is
so universal that the investigator has no difficulty in observ-
ing it. Quite on the contrary, his difficulty will be in getting
rid of the disturbing effects introduced by it.

External inhibition is a fairly obvious kind of effect which
one might have anticipated. The various types of internal
inhibition, however, are rather less obvious. First to be dis-
covered was the phenomenon which was to be known as that
of 'extinction'. Let us suppose that a conditioned salivary
response has been set up to a buzzer; let us now suppose that
the buzzer is sounded a number of times without reinforce-
ment, i.e. without being followed by the presentation of
food. Gradually the number of drops of saliva secreted will
diminish, until finally no saliva is produced at all. When this
stage is reached, the response is said to have been experi-
mentally extinguished, and superficially the situation now ap-
pears to be very much as it was before conditioning took

246 Part Two • Personality and Social Life

place. The previously neutral stimulus is neutral again in
the sense that it produces no reaction.

Why, it may be asked, do we require an added concept of
inhibition? Would it not be simpler to say that the animal
originally learned to respond with salivation owing to condi-
tioned stimulus and that now, owing to lack of reinforcement,
he has gradually forgotten to respond in this manner?
There are two reasons why this view is not acceptable. In
the first place, it requires an active process of extinction
to make the animal 'forget'. Conditioned responses not
actively extinguished are retained over periods of months
and years, so that the passive process of forgetting cannot be
appealed to in any reasonable manner.

What is more conclusive, however, is another pheno-
menon consequent upon extinction; namely, the pheno-
menon of recovery. Take again the dog who was conditioned
to the sound of a buzzer and whose conditioned response
was then extinguished to the point where no salivation what-
soever occurred to the buzzer. Test him again the next day,
and the conditioned response will again be present as strongly
as ever. It can again be extinguished, but after a reasonable
pause, recovery will once more take place. The conditioned
response, once it is firmly established, can be extinguished
dozens of times, but it will always return, like the proverbial
bad penny. These facts make it quite essential to conceive of
inhibition as an active process which is produced by experi-
mental extinction and which counteracts the effects of excita-
tion. The experiment suggests one important characteristic
of inhibition which follows inevitably from the facts given so
far. If the observed behaviour of the animal is conceived of
as the algebraic sum of so many units of excitation minus so
many units of inhibition, then we may say that when extinc-
tion is completed the amount of inhibition present in the
animal equals the amount of excitation. The next day, how-
ever, the animal, without any further training, makes a posi-
tive response again. This means that now there are more
units of excitation present than of inhibition. In other words,
inhibition dissipates more quickly than excitation. This very im-

Personality and Conditioning 247

portant property of inhibition we shall have ample oppor-
tunity of returning to later. For the time being we may put
down the sequence of events in a typical extinction experi-
ment in a figure which may serve as a graphical illustration
of what we have described so far (Fig. $a-b).

An important and useful extension of the concept of in-
hibition is that of disinhibition. We have already seen that
excitation can be inhibited by a disruptive external stimulus.
What would happen if such an inhibiting stimulus were to
be applied to an organism already in an inhibited state, such
as for instance a dog in whom the positive conditioned reflex
had just been extinguished? According to the theory, the
external inhibitory stimulus should inhibit the internal in-
hibition, and consequently the positive conditioned reflex
should again be produced. This is indeed what is found. Ex-
tinguish the conditioned reflex in a dog, then shine a bright
light into his eyes or ring a loud bell behind his head, and
quite suddenly the extinguished reflex will emerge again and
the dog will salivate to the conditioned (and extinguished)
stimulus. The experimental demonstration of this inhibition
adds one further argument to the acceptance of inhibition as
an active process.

We have quite vaguely mentioned that the most likely
locus for excitation lies in the modification of the structure of
the central nervous system, particularly at the synapses. It is
here also that we must look for the locus of inhibition. Little
is known about the precise anatomical details, but it may be
mentioned that in recent years it has been possible to detect,
under the microscope, physical changes at the synaptic level
corresponding to excitatory and inhibitory functions. It is
impossible as yet to say whether what has been observed
under the microscope is indeed the physiological basis,
directly or indirectly, of the psychological concepts of ex-
citation and inhibition. Much detailed work will be required
before any such identification can be made with any con-
fidence. However, even if this possibility should not stand up
to closer scrutiny, the value of a theory such as Pavlov's
would still be unimpaired. It serves to summarize existing

248

Part Two • Personality and Social Life

►

Atffil

Course of conditioned refkx -
Excitation - Inhibition

beginning of experimental
extinction

+8

+7-f

+6
+5 +
+4
+3
+2 i
+1
0 I
-1
-2
-3
-4
-5 ■
-6

-7
-8

Figure 5a: Hypothetical course of extinction experiment in which the
strength of the conditioned reflex is determined by the algebraic sum of
excitation and inhibition

knowledge, it serves to direct our attention to areas of signi-
ficance from the point of view of future research, and it tells
us precisely what to look for in our neurological and micro-
scopic studies. Few other theories in psychology can claim
as much.

Extinction, though probably the most widely studied
form of inhibition, is not the only one. Another important
variety is what Pavlov has called differential inhibition. In this
type of experimentation the animal is first conditioned to sali-
vate to a particular stimulus, such as for instance the sight
of a circle. When the animal is now shown an ellipse, we know
that because of stimulus generalizations he will again sali-
vate, although somewhat less than to the circle. If now we pro-
ceed to feed him every time he is shown the circle, but never
when he is shown the ellipse, his response to the ellipse will
be inhibited, so that after a while salivation will occur only
after the circle is shown and never after the ellipse is shown.

Personality and Conditioning

249

+8
+7 t
+6
+5
+4
+3
+2
+1
0
-1
-2
-3
-*
-5
-6
-7
-8}

Excitation

Recovery of extinguished reflex;
Strength = Excitation - Inhibition

1 2 3 4 5 6 7 8 9 10 1 12 13 Lf

Time in hours since extinction

-15^16 L7

Dissipation of Inhibition in time

Figure 56: Hypothetical course of recovery experiment in which the
recovery in the strength of the conditioned reflex is determined by the
dissipation of inhibition in time

This method of differential inhibition has been found im-
portant for two reasons. In the first place, by its use we can
learn something about the sensory acuity of dogs. Suppose
we want to know how much like a circle we can make an
ellipse while still keeping the two distinct from the point of
view of the observer. With human observers this presents no
difficulties, as we can rely on the verbally expressed judge-
ment of our subjects. But suppose we are interested in know-
ing the answer to this question in the case of dogs. In view of
their limited vocabulary we cannot use the same method as
with humans. What we can do, however, is to start out with
a conditioning experiment contrasting a circle with a rather
narrow ellipse, reinforcing the former and extinguishing the
latter. When the dog, through differential inhibition, has
succeeded in making this distinction, we take another ellipse,
slightly rounder than the first, and extinguish the dog's re-
sponses to this. We go on in this way, substituting ellipses

250 Part Two • Personality and Social Life

more and more like a circle, until finally, when we reach an
ellipse, the ratio of whose height to its width is about 8 to 7,
the dog's discriminatory capacity breaks down. This point,
then, marks his discrimination threshold for this particular
type of stimulus. We can, of course, repeat the experiment
with any other type of stimulus in which we may be inter-
ested, thus exploring in considerable detail the dog's ability
to make sensory discriminations.

This achievement, while important, is rather less signi-
ficant from the point of view of personality study than is an
observation made by Pavlov in the course of this experiment.
He discovered that when the breaking point was reached,
i.e. the point where the dog was unable to make the correct
differentiation, the dog then had what can only be described
as a kind of nervous breakdown. Its behaviour suddenly
changed in a variety of ways. It became extremely agitated,
started barking, refused to remain in the stand, became
aggressive to the experimenter, lost previously conditioned
reflexes, refused to go back into the experimental room, and
even refused food, although hungry. This unusual behaviour
might extend well outside the laboratory and even involve
such symptoms as sexual incapacity. Quite clearly we are
dealing here with a very unusual and profoundly important
type of reaction. We need not concern ourselves too much
with the rather semantic question of whether what Pavlov
had discovered was truly an 'experimental neurosis'; the
similarities certainly are striking, but we are so ignorant of
what precisely is involved in human neuroses that it is at
present impossible to consider Pavlov's discovery as any-
thing more than an analogy.

His theory regarding the origin of this experimental
neurosis certainly fits in with a conception of human neurosis
shared by many observers. In Pavlov's view the experi-
mental neurosis in his dogs was due to a conflict; a clash in
the cortex between excitatory and inhibitory potentials
elicited by the different stimuli which had acquired ex-
citatory and inhibitory properties respectively but which
at the perceptual level could no longer be distinguished

Personality and Conditioning 251

from each other. Physiologically it is difficult to see pre-
cisely what such a clash between excitatory and in-
hibitory potentials may mean. Psychologically the theory
certainly cannot be dismissed without serious considera-
tion.

This is so particularly because in another type of situa-
tion, also involving a clash between excitatory and inhibi-
tory potential, Pavlov was successful in also producing ex-
perimental neurosis. Certain strong and painful stimuli lead
to defensive reactions on the part of the animal and inhibit
the secretion of saliva and digestive activity altogether.
When a stimulus of this type, say a strong electric shock, is
used as a conditioning stimulus for salivation, we have a
clash between the normal reflex response to strong electric
shock (drying up of saliva) and the conditioning response
(salivation). Under these conditions, if Pavlov's theory is
right, experimental neurosis should occur again, and indeed
the facts bear out the hypothesis.

In these experiments we see the first link between the laws
of conditioning on the one hand and the experimental study
of personality on the other. Interestingly enough, Shaw,
who never seems to have got beyond the first chapter of
Pavlov's book, does not even mention these discoveries, pos-
sibly because even his black girl could not have told him
about them in twenty-five seconds from her abundant store
of common sense.

There are one or two other varieties of inhibition, but
these are rather less important and we need only mention
them in passing. One of them is so-called ' Conditioned In-
hibition'. It may be remembered that in elaborating a
secondary conditioned reflex, the new conditioned stimulus
is linked with the reflex through a conditioned stimulus pre-
viously elaborated, and that this is linked up by presenting
the animal with a new stimulus, turning it off again and let-
ting several seconds elapse before the previously conditioned
stimulus is applied. This lapse of several seconds between the
new and the old conditioned stimulus is absolutely essential
for the elaboration of higher order conditioned reflexes. If

252 Part Two • Personality and Social Life

the new stimulus is applied in conjunction with the old con-
ditioned stimulus, i.e. with no, or only a very short, time in-
terval between them, then the new conditioned stimulus
acquires inhibitory properties and is called by Pavlov a
'conditioned inhibitor'. These inhibitory properties can be
demonstrated by applying it simultaneously with some pre-
viously established conditioned stimulus. It will have the
effect of inhibiting the flow of saliva well below that which
would normally be forthcoming.

Another type of inhibition Pavlov has called the ' inhibi-
tion of delay5. We have already mentioned the existence of
trace reflexes in which the conditioned response does not
occur for a minute or two. Clearly, during this intervening
period, inhibition is present and suppresses the active emer-
gence of the reflex. That this view is the correct one can be
shown by the simple device of applying a disinhibiting
stimulus shortly after the conditioned stimulus has been
given. The animal will immediately respond with salivation
instead of waiting until the expiration of the time interval
to which the trace reflex had become conditioned.

One last type of inhibition must be mentioned. This is ' in-
hibition with reinforcement'. As Pavlov puts it 'the cortical
cells under the influence of the conditioned stimulus, always
tend to pass, though sometimes very slowly, into a state of
inhibition'. Pavlov was led to this belief by observing that
in certain circumstances it was not necessary to extinguish
the conditioned reflex along the lines described above. By
simply repeating the original conditioning procedure, i.e.
the presentation of the conditioned stimulus followed by the
presentation of the food, a few hundred times, the animal
would finally cease to show any kind of conditioned re-
sponse. This was a rather mysterious effect for which Pavlov
does not present a very convincing explanation. There seems
to be no doubt that inhibition with reinforcement does occur,
and there seems to be no doubt either as to the fact that the
conditioned stimulus does acquire inhibitory properties.
This can be shown in the same way as before, by pairing it
with another newly elaborated conditioned stimulus and

Personality and Conditioning 253

showing that it actively depresses the rate of salivation by
this newly elaborated stimulus.

Having now outlined very briefly some of the facts on
which Pavlov's theory of excitation and inhibition is based,
we may note one further theory of his which again links it
very closely with the theory of personality. While interested
in the general laws of cortical behaviour, Pavlov could not
help noticing very marked individual differences in the be-
haviour of the dogs with whom he was working. To take one
example, some dogs were easy to condition but rather diffi-
cult to extinguish; others formed conditioned reflexes only
with difficulty but were very quick to extinguish. These two
types of dogs also differed considerably in their general be-
haviour, and in terms of his theory, Pavlov considered these
differences due to innate properties of the central nervous
systems of these dogs. His explanation ran along the follow-
ing lines. In the average dog there is a certain balance be-
tween the development of excitatory and inhibitory poten-
tial. In some dogs the ratio of excitation over inhibition is
relatively high, and this disturbance of the balance leads to
their being easily conditioned and difficult to extinguish.
Conversely, in some dogs the ratio of excitation over inhibi-
tion is unusually low ; this leads to difficulties in conditioning
and to easy extinction (these consequences of a disturbance
in the excitation to inhibition ratio follow quite naturally
from Pavlov's conception of all excitation as responsible for
production of conditioned reflexes and of inhibition as being
responsible for their extinction).

In the last few years of his life, Pavlov became interested
in mental abnormality and spent a good deal of time in
psychiatric hospitals observing symptoms of mental patients.
He was struck with the apparent similarity of some of these
symptoms to the behaviour difficulties of his dogs and tried
to account for them in terms of his concepts of excitation and
inhibition. In doing so he followed Janet, a famous French
psychiatrist, by dividing neurotic patients into two great
groups; on the one hand there are the hysterics, people
characterized by a histrionic personality, a certain lack of

254 Part Two • Personality and Social Life

moral scruple, an overt interest in sexual matters, and con-
siderable liking for the society of other people. Symptoms
appearing in the more extreme forms of hysteria are para-
lyses, perceptual dysfunctions such as blindness or deafness,
and amnesias in which certain episodes or whole parts of the
individual's life might be forgotten (many of these symp-
toms, the reader may recall, were observed in Mesmer's
patients, and indeed nothing can demonstrate the lack of a
physiological basis of these hysterical symptoms more clearly
than the fact that they can be alleviated and often com-
pletely removed under hypnosis).

The other great group of disorders has been labelled in
many different ways at different times; perhaps the term
* dysthymic' describes them best. People in this group are shy
and unsociable in their behaviour, have strong emotions, are
given to anxiety and depression, and may even develop ob-
sessional and compulsive habits. Whereas the hysteric's
symptoms find an expression which is easily observable in
the outer world, those of the dysthymic are readily available
only to his own introspection. Everybody can observe the
paralysed limb or the functional blindness of the hysteric,
but the guilt-ridden anxiety and deep depression felt by the
dysthymic may often escape notice. Not all neurotics fall into
one of these two groups. Indeed, the majority would prob-
ably be found in a mixed group containing symptoms char-
acteristic of both hysterics and dysthymics. However, it is
possible in a rough-and-ready manner to arrange patients in
a continuum all the way from almost pure hysteria through
various mixtures to the other pole of almost pure dysthymia.

Pavlov was struck very much by the fact that the symp-
toms of the hysterics were always of an inhibitory nature.
Paralysis involved an inhibition of the motor-effector sys-
tem. Anaesthesias and other perceptual dysfunctions in-
volved an inhibition of the affector-perceptual mechanism.
Amnesias involve the inhibition of part of the cortical sys-
tems subserving memory. Conversely, the symptoms of a
dysthymic seemed to him to show evidence of an excess of
excitatory potential and a failure to develop sufficient in-

Personality and Conditioning 255

hibitory potential. This hypothesis, then, based upon these
observations and conjectures, runs as follows: hysterical
symptoms are developed by individuals in whom the
excitation-inhibition balance is tilted in the direction of ex-
cessive inhibition. Dysthymic symptoms develop in indivi-
duals in whom the excitation-inhibition balance is tilted in
the direction of excessive excitation.

Pavlov never followed this work up by actually putting
his theory to experimental test, but this has been done since
and the results support his view. It follows from his theory
that hysterics should be difficult to condition in view of the
excess of inhibitory over excitatory potential, whereas dys-
thymics should be very easy to condition in view of their
excess of over-excitative inhibitory potential. Several at-
tempts have been made to test this deduction and results
have always tended to support Pavlov's view.

The reader may be interested to know how conditioning
experiments are performed in human beings. While it is pos-
sible to use the salivary reflex, this is a little messy, and other
reflexes have usually been preferred. One of these is the eye-
blink reflex. Here the unconditioned stimulus is a puff of air
to the eyeball; the reflex associated with this stimulus is a
rapid closure of the eyelids. If now the conditioned stimulus,
which might be a tone delivered through a pair of earphones,
is made to precede the puff of air by about half a second a
number of times, then the eye-blink will become conditioned
to the tone, and occur even although no puff of air is de-
livered to the eyeball. Another method frequently used in-
volves the psycho-galvanic reflex, i.e. the sudden drop in the
resistance of the skin to an electric current which follows
any sudden stimulus and which we have described in an
earlier chapter. In the experiment the subject may be shown
a series of words on a screen. Every time one particular word
is shown to him an electric shock is administered. Very soon
the psycho-galvanic reflex which always accompanies the
shock becomes associated with the word itself, which thus
becomes a conditioned stimulus. Many other methods have
been used, but these are probably the most popular.

256 Part Two • Personality and Social Life

The relation between ease of conditionability and neuro-
tic symptomatology is certainly interesting, but it does not
seem to find any easy application to normal personalities.
The bridge is provided, however, by a well-known psychia-
trist - C.G.Jung. As is well known, Jung postulated the
existence of a continuum from extraversion to introversion
on which all human beings could be placed. From his ac-
count, the extravert emerges as a person who values the
outer world both in its material and in its immaterial aspects
(possessions, riches, power, prestige) ; he is sociable, makes
friends easily, and trusts other people. He shows outward
physical activities, while the introvert's activity is mainly in
the mental, intellectual sphere. He is changeable, likes
new things, new people, new impressions. His emotions are
easily aroused, but never very deeply. He is relatively in-
sensitive,impersonal, experimental, materialistic, and tough-
minded.

It may help the reader if this somewhat abstract descrip-
tion of the extravert and the introvert is rounded off with a
number of examples. These are taken quite at random from
real life and from fiction; the only thing they all have in
common is that all are good examples of strongly introverted
or extraverted people. In looking through the list the reader
must remember that both extraverts and introverts can be
bright or dull, stable or unstable, normal or mad. In saying
that two people are extraverted there is no implication that
they are alike with respect to all the various personality traits
which psychologists have discovered. They are alike only
with respect to those traits which form the syndrome or con-
stellation of traits which constitutes extraversion-intro-
version. If the reader will thus read quickly through the two
lists given below, trying to abstract what is common to all
the people in each list, he may obtain an intuitive under-
standing of the nature of extraversion and introversion.

Let us start out, then, with a baker's dozen of introverts,
followed by a similar number of extraverts. Here are the
introverts: Hamlet, Sherlock Holmes, Robespierre, Savo-
narola, Spinoza, Cassius, John Stuart Mill, the March Hare,

Personality and Conditioning 257

Sir Stafford Cripps, Faust, Gato the Elder, Don Quixote,
and Kant.

Next the team of extraverts: Mr Pickwick, Bulldog Drum-
mond, Boswell, Mr Punch, Caliban, Dumas, Donald Duck,
Churchill, Pepys, Cicero, Falstaff, and Toad of Toad Hall.

It would be wrong to credit Jung with the discovery of
this personality dimension. The very terms extraversion and
introversion can be found as far back as the sixteenth cen-
tury, and in more modern times the English psychologist
Jordan and the Austrian psychiatrist Gross both antici-
pated Jung in putting forward theories very similar to his.
He did, however, popularize this particular typology and he
made one important contribution to it. He pointed out that
hysterical disorders tend to develop in extraverted persons, whereas
introverts are more liable to symptoms of the dysthymic type. This
link-up of extraversion and hysteria on the one hand and
introversion and dysthymia on the other has received ample
experimental support. We can, therefore, extend Pavlov's
hypothesis and say that where the excitation-inhibition
balance is tilted in the direction of an excess of excitation we
are likely to find introverted individuals, whereas when the
balance is tilted in the opposite direction we are liable to
find extraverted individuals. According to this theory, then,
we would expect extraverts to be difficult to condition and
introverts to be easy to condition. This deduction also has
received experimental support.

In extending Pavlov's theory, however, we have lost the
connecting link between excitation and inhibition on the
one hand and personality on the other. While his hunch that
hysteria and excessive cortical inhibition were related was
merely based on reasoning by analogy, at least it did pro-
vide a link between these concepts. There appears to be no
direct link between inhibition and extraversion or between
excitation and introversion. To provide such a link will be
the last task of this chapter. In order to do so, however, we
must go on a slight theoretical detour.

In Pavlov's theory of conditioning what is required for
learning is mere contiguity. Pair the conditioned stimulus

258 Part Two • Personality and Social Life

with the unconditioned stimulus and the mere occurrence
together of these two stimuli will have the desired effect of
associating the conditioned stimulus with the reflex or re-
sponse. Yet this seems to run counter to a very basic prin-
ciple of human learning, the so-called law of effect, according
to which learning takes place only when a reward or a
punishment is provided. The typical experiment which
illustrates the law of effect is the so-called Skinner Box,
named after the well-known American psychologist who
first constructed it. Essentially it is a sound-proof box with a
glass lid, completely empty on the inside except for a small
movable lever near the bottom and a trough positioned near
the lever. The hungry rat is placed into this box; in the
course of its exploration it accidentally depresses the lever,
which is wired to a food magazine in such a way that a food
pellet is delivered into the trough. The rat quickly eats the
pellet and soon learns to press the lever in order to get
further pellets of food. The difference between the Skinner
experiment and the Pavlovian one is this: in Skinner's ex-
periment the learnt response, i.e. the lever pressing, is instru-
mental in providing the reward. In other words, the rat is
rewarded for making this particular response. In the Pav-
lovian experiment the learnt response, i.e. salivation, is not
in any way instrumental in producing the food ; the animal
would have obtained just as much food if it had never
learned to salivate to the conditioned stimulus. Hence
Skinner's type of experimentation is often referred to as 'in-
strumental conditioning', whereas the Pavlovian variety is
often referred to as 'classical conditioning'. One other
difference will be noticed: 'instrumental conditioning' deals
with conditioned responses involving muscles and bones,
whereas the responses involved in 'classical conditioning'
make use of glandular secretions and other activities asso-
ciated with the autonomic rather than with the central ner-
vous system. In a slightly different terminology, instru-
mental conditioning involves voluntary activities and classi-
cal conditioning involves largely involuntary activities.
Thus there is much to be said in favour of the view which

Personality and Conditioning 259

asserts that these two types of learning are differentiated in
many important ways and that the laws according to which
they operate are not by any means identical. There is no
point in going into this matter in any considerable detail
now, but we must notice how these two different types of
learning complement each other and contribute to the ac-
quisition of different kinds of skill by the human infant as he
grows up.

Roughly speaking, there are two different kinds of activi-
ties which the young child has to learn. The acquisition of
the first kind of activity is relatively easily accounted for in
terms of the law of effect. The young baby has to learn for
instance to suck at the mother's breast. He solves this prob-
lem very much as the rat solves the problem of the Skinner's
Box. Random movements of the head and mouth, perhaps
guided to some extent by the mother, produce milk and a
reduction in hunger, just as pressing the bar produced a
pellet of food in the case of the rat. A few repetitions enable
the infant to learn this series of events and to make use of this
instrumental conditioning for the satisfaction of his bodily
wants. This may serve as a prototype for the very many
different kinds of activities in which what the individual
learns benefits him directly and immediately. The law of
effect ensures the success of this type of instrumental con-
ditioning.

However, there are many activities which would be plea-
surable and rewarding in themselves, but which society can-
not permit. In the young child indiscriminate urination and
emptying of the bowels may serve as an example. In older
children and adults, we might use as an example the uncon-
trolled release of aggressive and sexual urges. The difficulty
of controlling what is often called one's 'animal nature' is
proverbial; the miracle is that it can be done at all. Instru-
mental conditioning and the law of effect do not help us
here; quite on the contrary, they would suggest that the
highly pleasant and stimulating consequences of satisfying
one's aggressive and sexual urges immediately and without
regard for the consequences should be learnt very firmly

260 Part Two • Personality and Social Life

indeed. How then is it possible for the process of socialization
to work? The answer is quite briefly that Pavlovian condi-
tioning is required as an additional variable. Unpleasant
autonomic responses such as pain and fear become condi-
tioned in the process of training to anti-social activities, and
the individual, by not indulging in these anti-social activities,
secures the immediate reward of a reduction in these painful
autonomic responses. This may be a difficult idea to digest
at first, and an example may make the conception more
readily intelligible.

Let us take the little brown bear. Like the human infant,
he also has to learn two types of activities : those which are
immediately beneficial to himself, and those on which society
has to insist as a condition of survival. As an example of the
first type of activity let us take the provision of food. The
mother bear has to teach him that blueberries are good to
eat. She has little difficulty in doing this; she simply picks
him up by the scruff of his neck, carries him to the nearest
blueberry bush, and dumps him into it. In the course of his
somewhat uncoordinated efforts to get out of the bush, he
accidentally squashes a few of the berries with his paws and
then reflexly licks his paws. The reward provided by the
taste of the blueberry juice ensures that through the agency
of instrumental conditioning he will from now on assidu-
ously hunt for blueberries.

But the mother bear has another much more difficult job.
The father bear, being somewhat cannibalistically inclined,
would like nothing better than to make a meal of his son.
The only way in which the mother can protect him is by
teaching him to climb up the nearest tree whenever she gives
him the signal that father is coming. She also has to teach
him to stay on the tree until she gives a signal that all is clear.
Now she can hardly explain these things to the little bear,
and she encounters the additional difficulty that he finds life
on the ground much more amusing than going up a tree and
staying there, very bored and very much against his will.

However, being, like most animals, a good psychologist,
she sets about her task very much according to the dictates

Personality and Conditioning 261

of Pavlovian conditioning. Picking up the little brown bear
by the scruff of his neck, she takes him to the nearest tree,
honks very loudly, and then gives him a painful bite on his
undercarriage. The little brown bear, surprised and hurt,
seeks to escape from his suddenly aggressive mother and
shins up the tree. After a while he tries to come down again,
but the mother rears up and gives him another painful nip
to send him upwards again. Finally, she gives two honks to
indicate that the trial is over and that the little bear may
come down again. The whole procedure is repeated a num-
ber of times, until finally the little bear has learnt his lesson
and the vigilance of the mother can protect him from the
baser instincts of his father. In fact, so well has he learnt his
lesson that when the mother bear finally decides that he is
old enough to fend for himself, she simply sends him up the
tree by giving the warning signal and then goes away and
leaves him for good. He is so well conditioned not to come
down without her permission that he will stay on the tree for
hours, even days, until hunger pains finally drive him down.
What has happened? By pairing the warning signal (the
conditioned stimulus) with the painful bite on his backside
(the unconditioned stimulus), the mother has set up a con-
ditioned reflex in which the warning signal produces a
powerful fear reaction on the part of the little bear ; a fear
reaction which can be relieved only by the action to which
he has become conditioned, namely that of shinning up the
tree. Thus the conditioned autonomic response becomes, as
it were, an intermediary in the law of effect; the reward
which the young bear gets for obeying the social laws of his
clan is a reduction in anxiety rather than any external re-
ward. Much the same happens when he is on top of the tree
and wants to come down before having received the all-
clear. As he begins to go down the tree, the stimuli he en-
counters have become conditioned in his past experience to
the powerful and painful bite received from his mother ; con-
sequently, the conditioned response, i.e. fear or anxiety, be-
come stronger and stronger until finally he climbs upwards
again in order to relieve his anxiety.

262 Part Two • Personality and Social Life

This account of the behaviour of the little brown bear
may appear a little fanciful, but there is good experimental
evidence in favour of the major theoretical points made
here. Admittedly, the experiments were carried out on rats
rather than on bears (even American departments are sel-
dom well enough equipped to be able to support a few hun-
dred bears in order to allow graduate students to complete
their Ph.D. theses). Nevertheless, the general outline of what
has been said is probably accurate enough to stand critical
scrutiny.

You thus arrive at the point where we consider condition-
ing the essential substratum of the socialization process.
Where the religious person talks about conscience as re-
straining the evil-doer, the psychologist would point to the
conditioning process as the agent responsible for the pre-
sence of the conscience in the mind of the evil-doer. In a
similar way, where Freud attributes unselfish and ethical
actions to a super-ego derived from parental teaching and
example, the psychologist would consider Pavlovian condi-
tioning as the method by means of which this goal is reached.
There is, in principle, no issue between the religious, the
Freudian, and the Pavlovian approaches. The main differ-
ence is that neither the Freudian nor the religious approach
provides an experimentally testable hypothesis to tell us the
precise method by means of which the final result of social-
ization is brought about. Perhaps Shaw's black girl could
have elaborated a testable theory in twenty-five seconds, but
her rather aggressive behaviour, denoting a strongly hysteri-
cal and extraverted personality, makes one doubt whether
she had much of a notion of the meaning of the socialization
process.

We are now in a position to link up, as we promised to do,
the conditioning process on the one hand and extraversion-
introversion on the other. We start out with the known fact
that there are marked individual differences in the excita-
tion-inhibition balance, differences which manifest them-
selves in different degrees of conditionability. Given that
some individuals are easier to condition than others, and

Personality and Conditioning 263

assuming for the moment that all individuals are subjected
to a similar process of socialization, it would follow from our
general theory that those who are most difficult to condition
should be relatively under-socialized, while those who are
relatively easy to condition would be, comparatively speak-
ing, over-socialized. Over-socialization and introversion
should therefore, in terms of our theory, go together, as
should under-socialization and extraversion. Is this really
so? The experimental evidence is not as extensive as one
might wish, but as far as it goes it definitely supports this
view. Let us look again at our neurotic extraverts and intro-
verts, the hysterical and dysthymic groups respectively, be-
cause in them we see to an exaggerated extent certain quali-
ties characteristic of extraverts and introverts altogether.
Close to the hysterics, but even more extremely extraverted
according to their test performances, are the so-called
psychopaths: these are people characterized by an almost
complete absence of social responsibility. Many of them are
pathological liars who tell lies almost by preference and re-
gardless of the certainty of being found out. Others commit
thefts without regard for the inevitable consequences;
others again go absent without leave or contravene other
rules and regulations pointlessly and in spite of the certainty
of being found out and punished. Psychopaths generally
seem almost completely lacking in this conscience or super-
ego, which is so essential in making civilized life possible.
Typically enough they also as a group are the most difficult
of all to condition and the most strongly extraverted in
terms of experimental tests.

Hysterics also tend to share the lack of a strong 'inner
light', as it were, which serves them as a guide to action.
They are easily swayed by momentary passions, by bad
companions, or by the standards of any small group of
which they happen to be members; while less extreme than
psychopaths, they also may rightly be considered under-
socialized.

Dysthymic groups show precisely opposite character-
istics. Where hysterics and psychopaths try to 'get away

264 Part Two • Personality and Social Life

with it' on every conceivable occasion, and even often under
conditions where detection is inevitable, the typical dys-
thymic not only does not indulge in anti-social activities, but
tends to worry excessively over the very slightest infringe-
ment of the social code which most people would dismiss
with a shrug of the shoulders. Where the psychopath might
seduce his girl-friend without a qualm, or commit bigamy in
order to gain his immediate ends, the typical dysthymic
would worry endlessly over ethical issues involved in a single
innocent kiss. Even relatively mild peccadilloes may lead
him to quite excessive methods of atonement, such as com-
pulsive hand-washing a hundred times a day to cleanse him-
self of some relatively unimportant misdoing. Small wonder,
then, that what characterizes the extravert most is to prefer
action to thought, whereas to the typical introvert, thought
is preferable to action. The stress of the socialization process
is largely on the inhibition of action; the abandoning of
aggressive or sexual activities of one kind or another. Conse-
quently, the introvert - the over-socialized person, who has
learnt his lesson too well - tends to generalize this rule to all
activity and prefers to seek salvation in his own thinking.
Conversely, the typical extravert, not having heeded the
lesson of the socialization process, prefers the immediate
satisfaction of his impulses through action.

This, then, is the general picture emerging from Pavlov-
ian theory and modern research. There are many z's to dot
and fs to cross before we can feel certain about the exact
relationships described here in broad detail, but it does not
seem likely that the main outline of the picture will require
any major revision. In any case, the substance of the chapter
will serve to show how personality measurement can be
geared to fundamental psychological theories which may be
far away from the more obvious type of personality test de-
scribed in the last chapter. It is in the further advance of such
more fundamental measurement that the greatest promise
for an increase in our knowledge of personality would seem
to lie.

7

POLITICS AND PERSONALITY

There are two contradictory points of view widely held
among many people nowadays, and frequently even held by
the same people at different times. Some believe that a per-
son's political opinions are the results of objective experi-
ences, of thought, and of definite decision; they are con-
sciously arrived at after a thorough weighing of the evidence
and are modifiable by logical argument and factual proof.
The opposite view is that political opinions are the reflec-
tions of personality, determined largely by irrational motives
of one kind or another, not amenable to logical argument or
factual disproof, and altogether an expression of personality
rather than a reaction to external reality. Many people feel
that the former of these two views adequately characterizes
the voting behaviour of themselves and their friends ; the
second type of motivation may be recognized more easily in
those voting for the opposite camp.

Among those who usually, or at least at certain times,
hold the view that personality factors are at least partly re-
sponsible for political views and social attitudes, there is
again a good deal of agreement on what are the basic traits
which are responsible for a person's choice of political party.
Their political opponents, so they declare, are driven to their
opinions by their lack of intelligence, their emotional in-
stability, and their chronic selfishness, which makes them
put class before country; conversely, those who think like
themselves are characterized by high intelligence, emotional
stability of an outstanding quality, and an imperturbable
integrity which makes them scorn prizes held out by rival
politicians.

Unlikely as such beliefs might appear, they have given
rise to a considerable amount of experimental work in the
social sciences, particularly in the United States. Quite a
large number of researches have been conducted on the

265

266 Part Two • Personality and Social Life

hypothesis that Socialists, who form a relatively small
minority group in America, are lacking in emotional stabil-
ity. The results, by and large, have failed to disclose any such
differences between Socialists and members of the Demo-
cratic and Republican parties respectively. Other studies,
involving the measurement of intelligence, have shown a
slight degree of superiority of the more radical as opposed to
the more conservative; this superiority, however, seems to
have been restricted to University students in the 1930's and
does not seem to apply to less highly selected samples at
different times during this century. Altogether, it may be
said that attempts, until quite recently, to link up personal-
ity and political beliefs have been restricted to the contrast
between conservative and radical opinions and have nearly
always resulted in failure. It is interesting to inquire into the
reasons for this failure and to show how, by more determined
application of the scientific method, success can be achieved
in this field also.

We may begin by taking account of two contradictory
types of statement which are frequently made and which,
being partial in their application, impede rather than
further the scientific study of social and psychological issues.
The first point of view to be mentioned here is one fre-
quently held by the more old-fashioned social scientists in
this country, by many politicians, and, implicitly at least, by
many groups of people who have only a tangential relation
to social science, such as historians, economists, and socio-
logists. This point of view might best be expressed in the fol-
lowing form. Empirical investigations in the social sciences
have a much lower and much less important status than
have philosophical arguments, scholarly reviews of opinions
held by well-known writers (preferably dead), and argu-
mentations about possible causes of historical events. Fac-
tual studies are looked upon with distaste because they force
a reconsideration of cherished values and beliefs, and be-
cause they do not usually fall in line with patterns of thought
formed many years ago.

This vague distaste for empirical work, which is quite

Politics and Personality 267

common in this country, has recently found a somewhat
more virulent expression in the United States. The 83rd
Congress established in 1953 a special committee to in-
vestigate Tax Exempt Foundations, under the chairman-
ship of Mr Rees. This committee collected evidence on the
support given by the Foundations to Social Science Re-
search, and made a special point of the criticism of empirical
investigations. The views expressed are so illogical and un-
clear that it is difficult to know precisely what is implied.
The type of insinuation made is exemplified by statements
like the following: 'It may not have occurred to (founda-
tion) trustees that the power to produce data in volume
might stimulate others to use it in an undisciplined fashion
without first checking it against principles discovered
through the deductive process.' If this statement means any-
thing - a point which is, of course, debatable - it means that
facts should not be discovered if they disagree with prin-
ciples discovered through the deductive process, i.e. if they
disagree with the a priori opinions of a person, or group of
persons, in charge of an inquiry, or conducting a Congres-
sional Committee investigation. This type of argument is,
of course, quite common in Russia (and in Hitlerite Ger-
many), where either no factual investigations in the social
sciences are carried out at all, or where the investigator is
told beforehand in no uncertain terms what kind of results
are expected from him. It is, however, a little disturbing to
find such views expressed in what is, effectively, an agency
of the Congress of a democratic nation.

It is easy to see why such distrust of factual empirical re-
search in the Social Sciences should arise among politicians
and others who have an axe to grind. Their success and their
very existence are predicated upon their ability to persuade a
large enough number of the population that the particular
beliefs they hold, and the panaceas they advocate, are in some
way advantageous to the community as a whole. They are
opposed by other politicians asserting precisely the opposite,
and, with the usual swing of the pendulum, both sides in due
course have their intoxicating draught of power given to

268 Part Two • Personality and Social Life

them by the electorate. They are well used to this particular
type of game and, by and large, have no ill feelings towards
their opponents.

The matter immediately becomes charged with emotion,
however, when the social scientist appears and says, 'Here
we have two opposite sets of hypotheses ; let us not waste
time in talking about which is nearer the truth, but let us go
and carry out an experiment to see which hypothesis is, in
fact, nearer the truth.' Such a proposal is almost invariably
considered by the politician as a threat to his particular
position in society. He has no means himself of carrying out
an investigation of this kind and, quite usually, he will even
be incapable of understanding the results if they are pre-
sented to him. He knows how to deal with a fellow politician
and with arguments he has heard a thousand times before,
but the party line will give him little support against the up-
start social scientist who takes these theories seriously and
actually wants to find out whether they work or not !

While it is thus easy to see why politicians should be
somewhat chary of employing the empirical approach, it is
difficult to see why the man in the street should have any
objections against it. As the President of the Social Science
Research Council has put it : ' To approach a problem em-
pirically is to say: "Let's have a look at the record." To
apply the empirical method is to try to get at the facts.
Where feasible, counting and measuring and testing are
undertaken. There is nothing necessarily technical about
empirical methods and there is no simple distinctive em-
pirical method as such. Congressional investigating com-
mittees normally follow an empirical approach. To imply
something immoral about using an empirical method of in-
quiry is like implying that it is evil to use syntax.' The alter-
native to an empirical study is speculation, aimless debate,
and unsupported theorizing. As John Locke, the famous
British philosopher, often called the father of empiricism,
said to a friend who mentioned to him some rationalistic
speculations by a Continental philosopher, ' You and I have
had enough of this kind of fiddling'.

Politics and Personality 269

In spite of the Rees Committee, then, we may conclude
that the social sciences, if they are to be anything at all ex-
cept idle speculation and arid, dry-as-dust scholasticism,
must have an empirical foundation; in other words, they
must be securely founded on ascertained facts. But is this
enough? Many social scientists seem to feel that the answer
here is 'Yes', and it is this second belief which I think re-
quires even more careful analysis and refutation than the
first. It is less obviously nonsensical but, none the less, it is
probably equally fatal to the development of a true science
of personality and social life. The reason for this belief is a
very simple one. Science is denned as systematic knowledge,
not just as knowledge, and while the empirical content is
certainly an absolutely essential part of it, the system, or
organization, of this empirical content is at least equally im-
portant. What the scientist is looking for is not a large num-
ber of disconnected facts ; it is rules or laws which bind to-
gether large groups of facts and make it possible, once a law
is known, to deduce these facts from it. Results have been
published in many different countries of thousands of
different Gallup Polls on all sorts of issues. These results pro-
vide a considerable amount of empirical content, but they
do not make Gallup Polling into a science. Only if it were
possible to find some general rule or law running through
all these results, and making it possible to deduce the in-
dividual results from the general law, would all this work
make a genuine contribution to science.

A simple example may illustrate this point. Hundreds of
thousands of stars have been discovered in the sky. All of
these are approximately circular in appearance, and there is
little doubt, therefore, that they are, in fact, spherical, or
very nearly so. Let us suppose that a new star has been dis-
covered. If we were asked to give an opinion as to its shape,
what would our answer be? Presumably, very few people
would fail to predict that this newly discovered star would
also be spherical, and the most common reason given for this
prediction would be that in the past all stars have been ob-
served to be spherical, and that the new one, presumably,

270 Part Two • Personality and Social Life

would not be an exception. This kind of argument does not
lack empirical content (after all, a large number of stars
have actually been observed in the past), but nevertheless it
is not a scientific argument. The scientist, when asked his
opinion, would make the same prediction as the man-in-the-
street, but he would make it on quite different grounds. He
would argue that, according to Newtonian laws of physics,
any large body made up of physical substances would, in the
course of time, assume a spherical shape. He would be able
to make this prediction even if no other stars had ever been
seen before, by a simple deduction from certain known laws.
It is the existence of such laws and the possibility of going
deductively from the law to the individual fact, and induc-
tively from a set of facts to a given law, that characterizes
science as opposed to the mere collection of isolated and
unrelated data.

Empirical work is at the foundation of science, but em-
pirical research is not enough. It must be guided by and
lead to theories of general significance and laws of deductive
power. Only under these conditions can we talk about a
field of study as being a science. As T. H. Huxley once put
this point in his pithy way, ' Those who refuse to go beyond
fact rarely get as far as fact ; and anyone who has studied the
history of science knows that almost every great step therein
has been made by the anticipation of nature; that is, by the
invention of a hypothesis which, though verifiable, often had
little foundation to start with ; and not infrequently, in spite
of a long career of usefulness, turned out to be wholly
erroneous in the long run.'

If this is so, and scientists, logicians, and philosophers of
science are in full agreement that these are some of the essen-
tial characteristics of scientific endeavour, can we consider
psychology and the social sciences as being truly scientific?
As a matter of fact, have they reached this stage, or are they
still in a pre-scientific stage of gathering isolated facts and
other unconsidered trifles? The answer, I think, must be
that some parts of psychology have already passed into the
scientific stage, others are still in a pre-scientific stage. To

Politics and Personality 271

those who would doubt the first part of this statement I
would like to offer the remainder of this chapter as an
example of the possibility of achieving deduction from
general laws and empirical verification of such deductions,
which we have just seen to constitute the essence of science.

The first thing that strikes us when we look at the field of
social attitudes, political behaviour, party strife, and voting
is that none of it appears to be innate in any sense, but that
all of it is due to some form of learning. However much it
may appear at certain moments as if the old jingle were true,
and 'Every boy and every girl that's born into this world
alive is either a little radical or else a little conservative', yet
in our heart of hearts we know that this is not so. Just
imagine the reactions of an Eskimo to an election campaign
largely waged in terms of the nationalization of the steel in-
dustry, or that of a Zulu, whose opinions are sought on the
relative importance of Federal as opposed to State rights !
We learn our politics as we learn our language, and if we
wish to know anything about political attitudes, then we
should be able to turn to the laws of learning in our efforts
for further clarification.

When we do this, we see that there appear to be two laws
rather than one. These two laws have been recognized for a
very long time indeed, although it is only in quite recent
years that they have been stated in a sufficiently clear form
to be amenable to experimentation. We might call these
laws the law of hedonism and the law of association. The
law of association simply states that we learn that A is fol-
lowed by B because in the past A and B have always, or
often, been associated with each other. The law of hedonism,
on the other hand, maintains that we learn things because
they have some effect on our well-being. Simple association
is not enough; it must be followed by some kind of reward or
punishment.

These two views may be summarized in terms of two ex-
periments. To characterize the associationist view we may
have recourse to Pavlov's dogs, which we discussed in a pre-
vious chapter. The simple association, repeated over a

272 Part Two • Personality and Social Life

number of times, of bell and meat causes the dog to learn that
the bell is followed by the meat, and to respond with saliva-
tion to the sound of the bell alone. In the Skinner box, on the
other hand, also mentioned in a previous chapter, the rat
produces a pellet of food by accidentally striking a bar in its
cage which is connected with a food reservoir; the relief
from hunger produced by this action causes the rat to learn
this particular movement whenever food is desired. These
two different types of modification of the animals' nervous
system as a result of experience may be called conditioning
and learning respectively.

We must now inquire into the consequences in the poli-
tical field which may be deduced from our knowledge of
these two different processes. Let us begin by noting certain
facts about the society in which we grew up, facts which
every youngster learns in the course of his early life. The first
fact we must know is that people differ with respect to their
social status. By status we mean such things as the amount of
money a man earns, the kind of education he has had, or
which he can afford for his children, the kind of house he
lives in, and the part of the town in which he lives, his accent,
the kind of people he mixes with, and so on and so forth.
At the one extreme we have the millionaire, who lives in a
huge house, in an exclusive quarter of the town, employs
several maids and butlers to look after him, runs several
cars, sends his children to Eton and Oxford, sports an old
school tie, and speaks with what everybody recognizes as a
* superior' type of accent. At the other end we have a down-
and-out, dozing on a bench in the Embankment, with no
one to look after him, shabby clothes, little food, and an
accent almost unintelligible to anyone not familiar with the
particular portion of the country he comes from. Between
these extremes there are all sorts of gradations, but, by and
large, there is little difficulty in fitting people along a con-
tinuum from the one to the other.

For practical purposes it is often useful to group people
into a number of status groups. There are many such classi-
fications. Typical of them is that used by the Gallup Poll.

Politics and Personality 273

Their top, or Av+ group, is characterized as follows: 'Well-
to-do men (or their wives) working in the higher profes-
sions, e.g. wealthier chartered accountants, lawyers, clergy-
men, doctors, professors, or in higher ranks of business, e.g.
owners, directors, senior members of large businesses. Al-
most invariably they will have a telephone, car, and some
domestic help.

'Av: Middle and upper middle class: Professional
workers not in the top category. Salaried clerical workers
such as bank clerks: qualified teachers: owners and mana-
gers of large shops : supervisory grades in factories who are
not manual workers: farmers, unless their farm is very big
when they will be Av+. Many will have a telephone, a car,
or employ a "char".

'Av— : Lower middle and working class: by far the big-
gest group. Manual workers, shop assistants, cinema atten-
dants, clerks, agents.

'Group D: Very poor: people without regular jobs or un-
skilled labourers or living solely on Old Age Pension. Hous-
ing will be poor. They can only afford necessities. '

Many psychological features are correlated with this
division of people into different status groups. The average
intelligence quotient of the Av-f- men (but not necessarily
their wives) would be 140-150; that of the Av group would
be in the neighbourhood of 120; that of the Av— group
would be slightly below 1 00 ; and that of Group D would be
around 90. I have discussed such relationships in some de-
tail in Uses and Abuses of Psychology and will not do so again
here. Suffice it to stress that the concept of status is not purely
defined in terms of positions, but is also related to psycho-
logical concepts.

In addition to status, which is an objective fact which can
easily be ascertained with regard to any particular person,
we have another concept which is much more subjective in
character, but which is also of considerable importance in
our analysis. That is the concept of social class. Whatever
their objective status may be, people in the democratic
countries tend to think of society as being grouped into

274 Part Two • Personality and Social Life

various classes, and they tend to consider themselves as be-
longing to one or other of them. This knowledge and this be-
lief can be seen to develop quite early in the lives of our
children ; by the time they leave school they are as well ac-
quainted with the concepts of class structure as are their
parents. While the concept of class is subjectively dependent
on each individual's private opinions and beliefs, it does, in
fact, have a strong factual relation to social status. The
Av+ group tends to think of itself as upper and upper-
middle class ; the Av group tends to think of itself as middle
class; while the Av — group, and more particularly the
very poor, tend to think of themselves as working class. The
relationship between social status and social class is shown
below in Table 3. Figures were obtained on a national
sample of about 9,000 people by the British Institute of
Public Opinion. It will be seen that social class (as estimated
in each case by the person interviewed for himself) and
social status (as estimated by the interviewer after talking to
the people concerned) do show considerable agreement. In
actual fact this Table underestimates the amount of agree-
ment which exists between the two concepts because the
interviewer's estimate of a person's status is known to be far
from completely reliable. When correction is made for that,
agreement becomes considerably higher than is shown in the
Table.

TABLE 3

Relationship Between Social Status and Social Class

Class

Upper and

Lower

Don't

Status

upper middle

Middle middle
(per cent)

Working

know

Av +

57

36 4

3

—

Av

16

58 13

10

3

Av-

2

20 20

55

3

Very poor —

7 8

76

11

We thus start our analysis with two widely known and
universally accepted facts, namely, that people differ with

Politics and Personality 275

respect to status and that they are aware of these status
differences, and as a consequence of them regard themselves
as belonging to certain social classes. We may go on by
noting that certain political issues may arise which will
further the ends of people belonging to one social class and
be opposed to the interests of people belonging to another
social class. Indeed, it would be difficult to find many issues
which do not in some way fall into this category. This well-
known truth is often put in the form of an analogy by refer-
ring to a national cake; however it is sliced, some people
will receive more than others and the interests of one group
will almost infallibly be opposed to the interests of another
group. Under these conditions, it seems almost inevitable
that political groups should arise to represent these respec-
tive interests, and, indeed, as is well known, groups of
parties have arisen in all the democratic countries which
represent this difference of interests. By long-established cus-
tom those parties representing the interests of the high status
groups are called conservative parties, or parties of the right,
while those parties representing the interests of the low
status groups are called radical parties, or parties of the left.
This bifurcation represents an inevitable consequence of
the law of learning and can be directly deduced from it. A
radical government, acting so as to further the interests of
the low-status groups, will thereby benefit members of the
low status groups, and thus reward them for having voted
for this particular party. Conversely, a conservative govern-
ment acting so as to further the interests of the high-status
groups will thereby benefit members of the high status
groups, and thus reward them for having voted for this par-
ticular party. A rat in a box which receives food when it
presses a red lever and an electric shock when it presses a
blue lever will very soon press the one and avoid the other.
Similarly, voters who receive benefits from one government
and have their status position lowered by another govern-
ment will soon learn to vote in accordance with their inter-
ests. There is nothing very mysterious or difficult about this
deduction, and, indeed, the general principle is fairly

276 Part Two • Personality and Social Life

universally accepted. Table 4 shows the relationship between
social status and political attitude in Great Britain; again, it
must be remembered that the relationship would be even
closer if estimates of social status were more reliable. Even
as the figures stand, however, there can be little doubt that
in a representative sample of the population there is a close
relationship between voting and social status. There is a
similar close relationship between social class and voting be-
haviour. Of those who consider themselves to be upper- or
upper middle-class, 79 per cent vote Conservative, whereas
only 20 per cent who consider themselves to be working
class do so. Only 5 per cent of the self-styled upper class
group vote Labour, whereas over 90 per cent of those who
consider themselves working- or lower middle-class do so.

table 4
Relationships Between Social Status and Political Attitude

Conser- Don't Total

Status vative Labour Liberal Other Know Number

(per cent)

Av -f- 77 8 11 — 3 447

Av 63 16 12 1 10 1,855

Av — 32 47 9 in 4,988

Very Poor 20 52 9 1 18 1,621

Total Number 3,411 3,545 894 60 1,001 8,911

It might be asked why the relationship is not perfect. If
our generalization is true and if the law is as stated, then
surely all working-class people should vote Labour, and ail
middle-class people Conservative. This is not so for a number
of reasons. In the first place, in our teaching a rat to press
one bar and avoid another, a large number of repetitions are
required of bar pressing behaviour. Conversely, the number
of elections in which a person takes part is relatively limited;
few people voting this year have voted in more than four or
five previous elections. It follows that the amount of re-
inforcement received is relatively small, and consequently a
good deal of random behaviour is to be expected.

Politics and Personality 277

Furthermore, in the case of the rat experiment, the re-
ward is inevitable and follows immediately. In the case of
political actions, the reward is not infallible and may not
follow immediately. At a given time the working class may
be better off under a Labour government than they would
be under a Conservative government at the same time, yet
because of world conditions outside the control of any
British government, their absolute well-being might be less
than that which they enjoyed under a previous Conservative
government. Under those conditions, the reinforcement pic-
ture is rather confused, and one would not expect a perfect
class alignment with voting behaviour.

Another point that arises is that voting behaviour may be
used to express feelings and emotions irrelevant to the issues
concerned. Thus, the son of a harsh and inconsiderate Con-
servative father may vote Labour, not because he feels any
affinity for the Socialist faith, but simply because he wants
to annoy his father. Conversely, the son of an equally harsh
Socialist father may vote Conservative for the same reason.
Causes of this type are not in any sense systematic and will
tend in the long run to cancel out. They do, however, serve
to make the correspondence between social class and status,
on the one hand, and voting behaviour on the other, to be in
less than perfect agreement.

It will now be clear why we would not expect, and, in-
deed, do not find any kind of correlation between personal-
ity and political behaviour as far as the right-left, or con-
servative-radical continuum is concerned. It may be possible
that very stupid people are slower to learn where their real
interests lie, thus perhaps voting against the party which
would, in fact, benefit them most. It is also possible that
some neurotic and emotionally unstable individuals may,
for obscure and irrational reasons, be opposed to the party
which best embodies their interests. There may be many other
individual features which in any given case make a particu-
lar person react in ways which are contrary to the general-
ization which we have established. However, these excep-
tions are not of a systematic kind ; the stupid working-class

278 Part Two • Personality and Social Life

person who might vote against his interests would be bal-
anced by the stupid middle-class person voting against his
interests, and thus there would be no correlation between
intelligence and tendency to vote for one party rather than
another. Thus, while these individual tendencies make the
law less than perfectly applicable, they do not produce any
kind of systematic tendencies.

One word should perhaps be said about the position in
the United States at the moment. It is often said there that
the country is relatively free from conceptions of social class,
and that the great political parties are not divided as Euro-
pean parties are on any class basis. Nothing could be further
from the truth. Numerous studies have shown that Ameri-
cans consider themselves members of the working-class or
the middle-class to pretty much the same extent as do people
in England, France, or Germany. It has also been found
that the political tie-up with class and status is quite strong
in the United States also. By and large, low-status groups
and people who consider themselves working-class tend to
vote for the Democrats, while higher-status groups and
people who consider themselves middle-class tend to vote
for the Republicans. The relationship between class and
status, on the one hand, and voting behaviour on the other,
is not quite as strong as it is in European countries, but it
has been increasing in strength over the years, and there is
every reason to believe that in a few years the party struc-
ture and party alignment in the United States will in every
way duplicate that observed over here.

We have now dealt in some detail with the consequences
to be expected from the hedonistic law of learning; we must
now consider the consequences to be derived through the
associationist law of conditioning. In doing this we must
have recourse to the argument already presented in an
earlier chapter. There it will be remembered it was shown
that the differences in a person's innate capacity to form
conditioned reflexes easily and quickly were responsible for
marked individual differences in temperament, particularly
along the extraversion-introversion continuum. We also

Politics and Personality 279

saw that the amount of socialization which a given person
succeeded in requiring was, to a great extent, determined by
his 'conditionability'. Thus, a person in whom conditioned
reflexes were formed easily and quickly will tend to become
1 over-socialized ' in comparison with the average, but a per-
son who forms conditioned reflexes slowly and with diffi-
culty will tend to become ' under-socialized ' in comparison
with the average.

In the fields of social behaviour and social attitudes the
two fields in which socialization should most clearly show
itself are the fields of sexual and aggressive behaviour. It is
here that we have the most obvious conflict between very
strong and powerful individual wishes and desires, and
equally strong and powerful social prohibitions and restric-
tions. Much the greater part of the process of socialization
may be said to consist in the erection of barriers to the
immediate satisfaction of aggressive and sexual impulses.
These barriers are absolutely essential if society is to survive
and in some form they exist even in the most primitive type
of society. Yet, however essential they may be to society,
they are irksome and annoying to the individual, who finds
himself thwarted in the expression of what, to him, are per-
fectly natural wishes and desires. Thus, here is a potential
area of great conflict, and it is here, if anywhere, that we
would expect the most marked contrast between the extra-
vert and the introvert; the easily conditioned and the poorly
conditioned. How would this conflict show itself in the field
of social attitudes and political behaviour?

Our expectation follows directly from what has been said
so far. We would expect to find a continuum ranging from
the introverted type of attitude to the extraverted type of
attitude, from the over-socialized to the under-socialized.
On the one side we would expect a strong insistence on bar-
riers of one kind or another to the free expression of sexual
and aggressive impulses. These barriers might be of a reli-
gious nature or an ethical nature, but what would be
common to all the beliefs and attitudes on this side would be
a desire to restrict the open expression of socially unaccept-

280 Part Two • Personality and Social Life

able behaviour. On the other side of the continuum, we
would expect to find the opposite, that is to say, a relatively
open demand for a relaxation of prohibitions, an overt de-
sire for the direct expression of sexual and aggressive urges,
and the denigration of religious and ethical standards felt to
stand in the way of such open manifestations. On the one
hand then, we should find a tender-minded regard for con-
ventions and rules protecting society from the more biologi-
cal drives of human nature; on the other hand, we should
find a tough-minded desire to over-ride these conventions
and seek direct expression of these animal instincts.

Thus, our hypothesis would lead us to predict the exist-
ence, in addition to our conservative-radical continuum, of
a tough-minded versus tender-minded continuum, quite un-
related to that of radicalism and conservatism, and there-
fore cutting clean across the more widely known and popu-
lar political schism. Before we turn to the evidence to see
whether, in fact, this prediction is borne out, let us first of all
look briefly at the structure of political groupings and
parties in this country. As Mr Disraeli said: 'Party is or-
ganized opinion5, and a study of attitude organization as
embodied in the major parties will therefore be relevant to
our problem.

In considering the mutual relations between the parties,
we find that two apparently contradictory theories are quite
widely held. The main groups in this country, Conserva-
tives, Socialists, Liberals, Communists, and Fascists, are
often thought to be arranged along a continuum from left to
right, so that the Communists are supposed to lie at the
extreme left, the Socialists a little nearer the centre, the
Liberals in the middle, the Conservatives to the right, and
Fascists at the extreme right. This theory is represented
graphically in Figure 6a.

On the other hand, many people find this arrangement
quite unconvincing. They argue that Communists and
Fascists have something in common which sets them off
against the democratic parties, and that to put them at
opposite ends of a continuum is manifestly absurd. Conse-

Politics and Personality 281

quently, it is argued, we should really have a different kind
of continuum, at the one end of which we would put Com-
munists and Fascists, at the other end the democratic
parties. Sometimes the argument is taken a little further,

Socialist

Cons&yative

(A)

Communist

Liberal

Tascist

Communist

1

Fascist

Socialist

(B)

•\Liberal

Conservative

Authoritarian

(c)

®
Communist

Socialist
l ©

Radical

Tascist

Conservative
©—

Conservative

i > Liberal

Democratic
Figure 6: Diagrammatic representation of three
hypotheses regarding the structure of attitudes as
mirrored in the relative positions of different poli-
tical parties

and it is said that the democratic parties also can be separ-
ated out on this continuum, with the Liberals being at the
opposite extreme to the Communists and Fascists, and the
Socialists and Conservatives a little less strongly opposed to
the tenets of these two authoritarian parties. This hypothesis
is shown diagrammatically in Figure 6b.

282 Part Two • Personality and Social Life

There is no doubt that most people would find some de-
gree of truth in both these hypotheses, contradictory to each
other as they may appear. This contradiction can easily be
resolved, however, if we decide that what is needed is not
one dimension or continuum but two, placed at right angles
to each other. One of these two continua, the radical-
conservative one, would sort out the parties in line from the
Communist at the left to the Fascist at the right; the other
continuum, which we may perhaps call the authoritarian-
democratic continuum, would lie at right angles to the first
and sort the parties out with the Communists and Fascists at
the authoritarian end and the Liberals at the democratic
end. This solution is shown in Figure 6c. It is presented at
the moment purely as a hypothesis and not as a factual
statement ; we will see later that the evidence in favour of a
hypothesis of this type is rather strong and that Figure 6c
does seem to represent reality to a considerable approxima-
tion.

We may go one step further and ask ourselves whether we
cannot identify this group of political parties with the organ-
ization of attitudes which we have deduced from the prin-
ciples of learning and conditioning. The radical-conserva-
tive continuum appears in both hypotheses and we may
readily accept the likelihood of our dealing with identical
concepts here. Can we go further and identify the authori-
tarian-democratic continuum with our hypothetical tender-
minded versus tough-minded continuum? On a priori
grounds, and judging from general knowledge and observa-
tion, there appears to be some justification for this. The high
degree of aggressiveness of the non-democratic parties in this
country is well known, as is also the extremely loose sexual
morality characterizing so many of the adherents of the two
extremist groups. (What is true of Communists in Europe,
Great Britain, and the United States does not necessarily
apply to the U.S.S.R. or other countries in which the Com-
munists have gained power.) However, common-sense judge-
ments of this kind have no scientific value, and while they
may make a particular solution appear reasonable, they can-

Politics and Personality 283

not serve as proof. Consequently, we must next turn to the
task of providing such proof, if, indeed, it can be found.

In looking for proof in favour of our hypothesis we must
first of all state a little more precisely what it is that we are
asserting. We are asserting that attitudes and opinions on
social issues are not independent of each other, but are or-
ganized and structured ; in other words, they tend to appear
in clusters. Furthermore, we are suggesting that there are
two main sets of clusters which together account for the
major part of the interrelations observed between different
attitudes. One of these clusters we have called the radical-
conservative one; the other cluster we have called the
tough-minded versus tender-minded one. What kind of evi-
dence can we adduce, first to show that attitudes are in fact
related to each other, and secondly to show that these rela-
tions give rise to the two sets of clusters we have specified?

The answer to the first part of this question is a relatively
easy one, and it depends essentially on the statistical method
of implication. Let us take two logically independent attitude
statements. For instance, 'Jews are cowards' and 'The Jews
have too much power and influence in this country'. Logi-
cally there is no relationship between these two statements.
A person may be a coward without having much power and
influence in the country, and, conversely, a person may have
a good deal of power and influence without being a coward.
Therefore, logically, there is no indication that a person
endorsing one statement should also endorse the other.
We can, however, posit that there is a continuum of anti-
Semitism such that some people will tend to endorse all anti-
Semitic statements while others will endorse none of them.
If that were true, then we would find that, in fact, people
would tend either to endorse both statements or to endorse
neither; relatively few people would be found to endorse
one but not the other. Let us suppose that we have inter-
viewed a thousand people and that, as indicated below, 450
have endorsed both statements, 350 have endorsed neither
statement, 100 have endorsed the statement that Jews are
cowards, but not that they have too much power and in-

284 Port Two • Personality and Social Life

fluence in this country, while another 100 have endorsed the
statement that Jews have too much power and influence,
but not the one that they are cowards (these figures are
quite fictitious and put in merely for the purpose of illustra-
tion) .

The Jews have too much power
and influence

YES NO

The Jews are cowards yes 450 100

NO IOO 350

It will be seen from this table that there is, indeed, a fac-
tual implication from one statement to another. The person
who believes that Jews are cowards is four and a half times
as likely to believe that Jews have too much power and in-
fluence in this country as a person who does not believe
that Jews are cowards. We can measure the strength of
implication existing in a table of this kind and express it in
the form of a single figure. This figure is usually referred to
as a coefficient of correlation, and varies from zero, when
there is no implication at all, to 1 -oo, when there is perfect
agreement. If the number of people in each of the four cells
of our table had been exactly 250 there would have been no
implication, and the correlation would have been zero. If
the number of cases in the cells agreeing with both state-
ments, had been 500, and 500 disagreeing with both state-
ments, then the implication would have been complete and
the correlation would have been 1 -oo. Knowing the person's
opinion on one issue would have determined completely his
opinion on the other issue.

In actual fact correlations between different attitudes
usually range from about 0-2 at the lower end, to about 07
or o-8 at the higher end. Thus, there is a definite amount of
implication in the field of social attitudes, but these implica-
tions represent tendencies, not certainties.

When we study, by statistical methods, the actual impli-
cations found in large samples of the population we find
that they are organized in a kind of hierarchical system. At

Politics and Personality

285

the bottom of the hierarchy we find a million and one casual
expressions of attitude or opinion which we make in the
course of our lives. Some of these are characteristic of our
long-range views, others are purely ephemeral and may just
be the outcome of a temporary annoyance. Thus, the driver
who has his new car scratched by an incompetent lady who
drives in a rather haphazard manner may be moved to call
out some deprecatory remark concerning woman drivers,
without, in his more sober moments, necessarily endorsing
the anti-feminist position implied. It is only when opinion
statements are made on more than one occasion that we

Ideology levd

Conservatism

Attitude level

Habitual
oplnidh levd

Specific
cfiruon O
levd

000 00 06006 66 66066 66600
Figure 7: Diagrammatic representation of the structure of attitudes

reach a relative stability of opinion which makes it worth-
while to measure and record the expression. Thus, if a per-
son on several occasions gives it as his opinion that children
should be seen and not heard, then we may regard this as a
genuine expression of opinion.

Such opinions themselves are intercorrelated ; we have
seen an example in the case of the two views that 'Jews are
cowards' and that 'Jews have too much power and influence
in this country'. Relationships such as these, which involve
a large number of opinions regarding one central issue (atti-
tude towards the Jews in this case) give rise to a somewhat
higher-order concept than that of opinion, namely the

286

Part Two • Personality and Social Life

concept of attitude. Attitudes themselves, however, are not
unrelated, and when we analyse their relationships we come
to a higher-order construct yet, namely, that of an ideology,
such as, for instance, the ideology of conservatism.

A hierarchical system such as this is illustrated in Figure 7.
In viewing this figure, it should be remembered that the re-
lationships implied therein are not in any sense arbitrary, or
based on a priori considerations on the part of the investiga-
tor. The Figure simply represents in a diagrammatic form

Companionate ^=
marriage^

Easier divorce laws &n

Sunday observance® ^ —
old-fashioned , ^^^
Abolish abortion^
& licensing laws

4-®-

Removemarriaqebar ~S
onjemak^ieachers

RADICALISM

■*

V

^coburedpeople iiiferior

^jewstoo powerful
®¥bgging

Harsh

X:
o-
d:
g^ 'Spare theroi'

CONSERVATISM

® ®Anti-misce<genation

Nationalization
inefficient

Make religious education
® compulsory

39) Make birth-control
=^ ill

Abolish ®?atriotismjvrce^.

private property against peace zffi r

No compulsory /^

nSd® sterilization^

sovereignty ®Pacij\sm .
COS not traitors &d=z
s

£L

A-

llegal
^Gobackto religion
\

-\

Figure 8 : Relationship of different social attitudes to each other and to
the two great principles of organization as determined empirically

actual relationships observed between attitudes and opinions
held by representative samples of the population. The im-
plication from one opinion or attitude to another is a factual
one; its measurement is based on the expressions of points of
view made by thousands of people who are selected at ran-
dom from the general population. This is an important
point to remember. What the psychologist contributes is a
theory and a method of analysing and organizing the data.
The actual content of this scheme, however, is contributed
by the people whose opinions and attitudes he is analysing.

Politics and Personality 287

His hypothesis may call for a high correlation between two
opinions, but whether such a high correlation does or does
not obtain is a factual question which only empirical re-
search can answer.

What, then, is the verdict of empirical research of this
kind when applied to the hypotheses we are considering at
the moment? The result of carrying out detailed analyses of
this kind on samples of several thousand men and women,
both working class and middle class of all degrees of educa-
tion, of all ages, and voting for all the different political
parties in the country, is shown in Figure 8. We can see
there the organization of opinions and attitudes, as objec-
tively determined by the calculus of implication, and it will
be seen that the outcome strongly supports our general
theory. There are, indeed, two main opposed sets of ideo-
logies corresponding to the radicalism and conservatism
and the tough-minded and tender-minded one respectively.
Conservative beliefs apparently include the view that
nationalization is inefficient; that religious education should
be made compulsory; that coloured people are inferior; that
birth control should be made illegal; that the death penalty
should be retained; and so forth. Radicals, on the other
hand, believe that private property should be abolished;
that Sunday observance is old-fashioned; that we should
give up our national sovereignty in the interests of peace; and
so forth. That these items do, in fact, represent conservative
and radical opinion can be shown by a very simple calcula-
tion. The percentage of endorsements for each item was cal-
culated separately for Conservative and Labour party
voters who were equated for social class, education, age, sex,
and other important variables. Then the difference in the
percentage of endorsements was calculated for each item
and it was found that the views mentioned above, as well as
the others given in the Figure, did in fact show very marked
differences in endorsement between Conservative party
voters and Labour party voters. There seems to be no
doubt, therefore, that we are here dealing with a genuine
radical-conservative continuum.

288 Part Two • Personality and Social Life

How about the tough-minded-tender-minded con-
tinuum? It will be seen that this bears out very strongly our
prediction. On the tough-minded side we have openly
aggressive and sexual attitudes. The aggressive ones favour
flogging, the death penalty, harsh treatment for criminals;
corporal punishment for children; and so forth. The openly
sexual attitudes are those in favour of companionate or trial
marriage; easier divorce laws; the abolition of abortion
laws, thus making abortion easily available to everyone ; and
so on. On the other hand, attitudes characterizing the
tender-minded end of the continuum emphasize ethical and
religious restraints and pacifism; the giving-up of national
sovereignty; going back to religion; and making religious
education compulsory ; the making illegal of birth control
and the abolition of flogging and the death penalty - these
are characteristic of tender-minded views. There appears to
be, therefore, considerable evidence in this analysis of atti-
tudes that our deduction is verified.

We have not, however, shown so far that our analysis of
the relationships between political parties is as predicted in
Figure 6c, and to this task we must now turn. If our hypo-
thesis is correct, then we would expect Communists to be
tough-minded radicals; Fascists to be tough-minded con-
servatives; Liberals to be tender- minded and intermediate
with respect to the radicalism-conservatism continuum;
Socialists and Conservatives should be intermediate with re-
spect to tender-mindedness between the authoritarian
parties and the Liberals, and to the left and right, respec-
tively, on the radicalism-conservative continuum. Proof of
this prediction was sought in the following way. Question-
naires were made up of items which best characterized the
conservative-radical continuum and the tough-minded-
tender-minded continuum; these two questionnaires we
shall call the R inventory and the T inventory respectively.
These inventories may be considered as reasonably reliable
and valid measures of our two continua. They were tried out
and refined in a number of studies, and finally applied to
large groups of people who were members of, or had voted

Politics and Personality 289

for, the various party groups with whom we are dealing.
Their scores on these two inventories were then determined
and plotted. When this was done it was found that our pre-
diction of the relative positions of these various groups were
borne out in very considerable detail. The only discrepancy
occurred with respect to the Fascist group, which was indeed
tough-minded and Conservative, but slightly less conserva-
tive than voters for the Conservative party. In all other de-
tails, the prediction was precisely fulfilled, and we may
therefore conclude that our hypothesis possesses a consider-
able degree of predictive value.

The reader may be interested in assessing his own and
other people's position on these two continua. For this pur-
pose I have given below a 60-item inventory which can be
scored for both the R and the T continuum. The reader
may like to fill this in first before looking at the key. The in-
structions preceding the inventory are those normally given
with it. The key to this, as well as the method of scoring and
the average scores for various representative samples, are
given at the end of this chapter.

SOCIAL ATTITUDE INVENTORY

Below are given sixty statements which represent widely-held
opinions on various social questions, selected from speeches, books,
newspapers, and other sources. They were chosen in such a way
that most people are likely to agree with some, and to disagree with
others.

After each statement, you are requested to record your personal
opinion regarding it. You should use the following system of
marking :

-\ — 1— if you strongly agree with the statement
-f- if you agree on the whole

o if you can't decide for or against, or if you think the ques-
tion is worded in such a way that you can't give an answer
— if you disagree on the whole
if you strongly disagree

Please answer frankly. Remember this is not a test ; there are no
'right' or 'wrong' answers. The answer required is your own per-

290 Part Two • Personality and Social Life

sonal opinion. Be sure not to omit any questions. The questionnaire
is anonymous, so please do not sign your name.

Do not consult any other person while you are giving your
answers.

Opinion Statements Tour Opinion

1. The nation exists for the benefit of the indivi-
duals composing it, not the individuals for the
benefit of the nation.

2. Coloured people are innately inferior to white
people.

3. War is inherent in human nature.

4. Ultimately, private property should be abol-
ished and complete Socialism introduced.

5. Persons with serious hereditary defects and
diseases should be compulsorily sterilized.

6. In the interests of peace, we must give up part
of our national sovereignty.

7. Production and trade should be free from
government interference.

8. Divorce laws should be altered to make divorce
easier.

9. The so-called underdog deserves little sym-
pathy or help from successful people.

10. Crimes of violence should be punished by
flogging.

1 1 . The nationalization of the great industries is
likely to lead to inefficiency, bureaucracy, and
stagnation.

12. Men and women have the right to find out
whether they are sexually suited before mar-
riage (e.g. by trial marriage).

13. 'My country right or wrong' is a saying which
expresses a fundamentally desirable attitude.

14. The average man can live a good enough life
without religion.

15. It would be a mistake to have coloured people
as foremen over whites.

16. People should realize that their greatest obliga-
tion is to their family.

1 7. There is no survival of any kind after death.

Politics and Personality 291

Opinion Statements - contd Your Opinion

18. The death penalty is barbaric, and should be
abolished.

19. There may be a few exceptions, but in general,
Jews are pretty much alike.

20. The dropping of the first atom bomb on a
Japanese city, killing thousands of innocent
women and children, was morally wrong and
incompatible with our kind of civilization.

2 1 . Birth control, except when recommended by a
doctor, should be made illegal.

22. People suffering from incurable diseases should
have the choice of being put painlessly to death.

23. Sunday-observance is old-fashioned, and should
cease to govern our behaviour.

24. Capitalism is immoral because it exploits the
worker by failing to give him full value for his
productive labour.

25. We should believe without question all that we
are taught by the Church.

26. A person should be free to take his own life, if
he wishes to do so, without any interference
from society.

27. Free love between men and women should be
encouraged as a means towards mental and
physical health.

28. Compulsory military training in peace-time is
essential for the survival of this country.

29. Sex crimes such as rape and attacks on chil-
dren, deserve more than mere imprisonment;
such criminals ought to be flogged or worse.

30. A white lie is often a good thing.

31. The idea of God is an invention of the human
mind.

32. It is wrong that men should be permitted
greater sexual freedom than women by society.

33. The Church should attempt to increase its in-
fluence on the life of the nation.

34. Conscientious objectors are traitors to their
country, and should be treated accordingly.

35. The laws against abortion should be abolished.

292 Part Two • Personality and Social Life

Opinion Statements - contd Tour Opinion

36. Most religious people are hypocrites.

37. Sex relations except in marriage are always
wrong.

38. European refugees should be left to fend for
themselves.

39. Only by going back to religion can civilization
hope to survive.

40. It is wrong to punish a man if he helps another
country because he prefers it to his own.

41. It is just as well that the struggle of life tends to
weed out those who cannot stand the pace.

42. In taking part in any form of world organiza-
tion, this country should make certain that none
of its independence and power is lost.

43. Nowadays, more and more people are prying
into matters which do not concern them.

44. All forms of discrimination against the coloured
races, the Jews, etc., should be made illegal, and
subject to heavy penalties.

45. It is right and proper that religious education in
schools should be compulsory.

46. Jews are as valuable citizens as any other group.

47. Our treatment of criminals is too harsh; we
should try to cure them, not punish them.

48. The Church is the main bulwark opposing the
evil trends in modern society.

49. There is no harm in travelling occasionally
without a ticket, if you can get away with it.

50. The Japanese are by nature a cruel people.

51. Life is so short that a man is justified in enjoy-
ing himself as much as he can.

52. An occupation by a foreign power is better than
war.

53. Christ was divine, wholly or partly in a sense
different from other men.

54. It would be best to keep coloured people in
their own districts and schools, in order to pre-
vent too much contact with whites.

55. Homosexuals are hardly better than criminals,
and ought to be severely punished.

Politics and Personality 293

Opinion Statements - contd Tour Opinion

56. The universe was created by God.

57. Blood sports, like fox hunting for instance, are
vicious and cruel, and should be forbidden.

58. The maintenance of internal order within the
nation is more important than ensuring that
there is complete freedom for all.

59. Every person should have complete faith in,
some supernatural power whose decisions he
obeys without question.

60. The practical man is of more use to society than
the thinker.

There are one or two more points which should be dis-
cussed. The first of these relates to class differences in tough-
mindedness. We have seen that from our hedonistic learning
field we can predict that working-class groups should be pre-
dominantly radical in their sympathies, while middle-class
groups should be predominantly conservative. Can we make
any prediction about class differences from our association-
ist conditioning hypothesis? The answer appears to be in the
affirmative. So far we have assumed in our deduction that
people who condition easily and people who condition with
difficulty are also submitted to a socialization process which
is roughly equal in strength or superiority for all of them.
This, however, is surely not true to the facts. We know that
some children are submitted to a very strict process of
socialization, others to a very lax one indeed. The outcome,
undoubtedly, will be determined not only by the degree of
conditionability of the child, but also of the amount of con-
ditioning to which he is subjected. Given an equal degree of
conditionability in a group of children, we would expect
those to be most 'over-socialized' who have been subjected
to a very strict socialization process and those to be ' under-
socialized' who have been submitted to a very lax process of
socialization.

Now, there is no reason to assume any differences between
social classes with respect to conditionability, but there are

294 Part Two * Personality and Social Life

very good reasons for assuming considerable differences be-
tween them with respect to the degree of socialization to
which they are subjected. Particular attention has been
drawn, for instance, by Kinsey in the United States to the
different value laid on the repression of overt sexual urges
by middle-class and working-class groups. He has shown
that where, for middle-class groups, parents put very strict
obstacles in the way of overt sexual satisfaction of their grow-
ing children, and inculcate a very high degree of ' socializa-
tion' in them, working-class parents, on the whole, are much
more lax and unconcerned. In many working-class groups,
for instance, he found pre-marital intercourse viewed as not
only inevitable, but as quite acceptable to the group.

Similarly, with respect to aggression, there is a consider-
able amount of evidence from a variety of sociological
studies, carried out both in the United States and in Great
Britain, to show a tendency for middle-class groups to im-
pose a stricter standard upon their children than the work-
ing-class groups. The open expression of aggressiveness,
which is frowned upon in the middle-class family, is often
not only accepted but even praised in the working-class
group.

There are, of course, many individual exceptions which
will immediately occur to the reader. There are middle-class
families where parents completely fail to impress social
mores and customs upon their children, and where the open
expression of anti-social tendencies is openly condoned, or at
least not strongly discouraged. Conversely, in many work-
ing-class families, particularly where there are ambitions
for the children to move in the direction of the higher-status
groups, there is an extremely strong tendency to stress con-
ventional values and inculcate respect for them in the chil-
dren. However, in spite of numerous exceptions, available
data leave little doubt that, on the average, there are differ-
ences between social classes and status groups which, by and
large, allow one to generalize in the direction of stating that
the socialization process is stronger and more complete in
middle-class than in working-class groups.

Politics and Personality 295

If that is so, and if we have no reason to expect innate
differences in conditionability between the two groups, then
we would expect our middle-class groups to be more tender-
minded than our working-class groups. This hypothesis has
been put to the test by applying the T-inventory to groups
of working-class and middle-class individuals, matched on a
number of relevant variables. The outcome was exceedingly
clear-cut. Middle-class Conservatives were more tender-
minded than working-class Conservatives; middle-class
Liberals were more tender-minded than working-class
Liberals; middle-class Socialists were more tender-minded
than working-class Socialists. Indeed, even middle-class
Communists were found to be more tender-minded than
working-class Communists! Not enough Fascists were
available to carry out a comparable study with them, but
for all the other groups considerable differences were found
in the predicted direction, and we may therefore generalize
and say that middle-class people, on the whole, tend to be
more tender-minded, working-class people on the whole
more tough-minded. There is, of course, considerable over-
lap, but of the existence of a noticeable average difference
there can be no doubt.

Another question that may concern us is that of national
differences. Most of the work mentioned has been carried
out in Great Britain, and it does not necessarily follow that
what is true of Great Britain is true of other countries. For-
tunately, a number of studies are available to show that the
organization of attitudes in France, Germany, Sweden, and
the United States is very similar indeed to that found in this
country. Similarly, the relationships of the R and T dimen-
sions to political parties are similar to those observed here.

There are, however, certain differences which in them-
selves are very enlightening; thus, in Great Britain the major
political parties are differentiated almost entirely in terms
of the radicalism-conservatism continuum, and compara-
tively little with respect to tough-mindedness and tender-
mindedness. Socialists and Conservatives are approximately
equal with respect to their degree of tough-mindedness; the

296 Part Two • Personality and Social Life

Liberals are a little more tender-minded, but the difference
is not very large. It is only the minority groups, like Com-
munists and Fascists, which show considerable deviation
from the norm.

When we turn our attention to France, however, the
position is very different. There it has been found that the
T-dimension, very far from being negligible, is, on the con-
trary, even more important than the radicalism-conserva-
tism dimension. Whereas in Great Britain the ratio of
importance for these two factors is approximately i o to i in
favour of radicalism-conservatism, in so far as the major
political parties are concerned, it is 4 to 3 in favour of
tough-mindedness-tender-mindedness in France. (This
might, indeed, have been expected in view of the well-
known strength of the Communist and of various Fascist
groups on the French political scene.)

This finding is very important to anyone who wishes to
compare the political structure in England and France. In
this country parties are divided from each other in terms of
radicalism-conservatism; the division between tough-
minded and tender-minded usually divides each of the
major parties into sub-sections. In France, on the contrary,
the major divisions are with respect to tough-mindedness-
tender-mindedness, and the radical-conservative dicho-
tomy is frequently found within each of the major parties.
The position is not so clear in France as it is in this country
because there the two principles, or dimensions, are of ap-
proximately equal strength. Nevertheless, a tendency is
there, and it will be apparent now why it is so very difficult
for British observers, used to our type of political organiza-
tion, to understand the quite different pattern on the French
political scene. If the reader, in studying French politics,
will bear in mind these considerations, I think he will find
his task considerably eased and his understanding grow
correspondingly.

What about countries and nations outside the European
circle of circumstance and background ? We may expect the
tough-minded-tender-minded dimension to remain in some

Politics and Personality 297

form because it is grounded on relatively universal and per-
manent characteristics of human beings. We would not,
however, expect to find the radical-conservative dimension
to emerge in the same form as here unless socia1 conditions
have given rise to classes and status groups of a similar
nature to those dominating the social scene in the European
and North American countries. In a feudal society, for in-
stance, you would not expect to find anything resembling
our conservative-radical continuum to emerge.

Only one study has been carried out along these lines in a
semi-feudal country. Among mid-Eastern Arabs it has been
found that while the tough-minded-tender-minded dimen-
sion is still clearly expressed in the relationships observed be-
tween different attitudes there is nothing that corresponds
to the radical-conservative continuum. It could be of the
very greatest interest if studies of this kind could be carried
out in countries like China, Russia, and so forth, but the
practical difficulties and the financial outlay involved make
it unlikely that in the near future our knowledge will grow
by an inclusion of these countries in our circle of exploration.

It will have been noticed that nothing has so far been said
of an implication in our theory which would link personality
with tough-minded and tender-minded attitudes respec-
tively. We have argued, on the one hand, that a person diffi-
cult to condition should develop tough-minded attitudes.
In a previous chapter we have argued that a person difficult
to condition should develop extraverted patterns of be-
haviour. Similarly, we have argued in this chapter that per-
sons particularly easy to condition should develop tender-
minded attitudes, and in a previous chapter we have argued
that such a person should develop introverted behaviour
patterns. It seems reasonable to expect, therefore, that a
tough-minded person would tend to be extraverted, and a
tender-minded person would tend to be introverted. This
hypothesis has been put to the test several times, and results
in each case have strongly supported this hypothesis. There
appears to be a definite tendency for extraverted people to
develop tough-minded attitudes, whereas introverted people

298 Part Two • Personality and Social Life

show an equally definite tendency to develop tender-
minded attitudes. With this finding our set of hypotheses has
indeed come full circle, and we now see in detail the relation-
ships obtaining between personality, social attitudes, and
political action. Considering the many sources of error in-
volved in the measurement and determination of personal-
ity and of social attitudes, the relationships found are re-
markably close. Nevertheless it should be remembered that
they are not perfect. We have found that Communists tend
to be extraverted, tough-minded radicals, whereas Fascists
tend to be extraverted, tough-minded conservatives. ^ It
should not be deduced from that that the converse also holds
and that all extraverted, tough-minded radicals are Com-
munists and all extraverted, tough-minded conservatives are
Fascists. All policemen are over 6 feet tall, but not all men
over 6 feet tall are policemen; the argument from the one to
the other cannot be reversed. (I understand that regulations
about the height of policemen, like most other things, has
been subject to change and nowadays one can behold police-
men who are only five feet nine inches tall. I have attempted
to obtain empirical evidence on this point, but have been
awed by the curious headgear worn by policemen which
makes any measurement of their height extremely un-
reliable.)

Indeed, it would not even be true to say that all members
of the Communist party, say, are in fact tough-minded radi-
cals. People join a party for all sorts of reasons and it would
be unreasonable to expect perfect conformity. To take an
extreme case, an agent provocateur, or police spy, might
join an extremist party in order to keep an eye on it; you
would not expect him necessarily to share its views and
attitudes. Again, we have found in several cases that a hus-
band belonging to, say, the Communist party is made to
persuade his wife to join also; she may join because she does
not want to break up her marriage, but without in fact
sharing the beliefs and aims of the party. There will always
be a number of people in a party whose membership is based
on considerations more or less tangential to the views held

Politics and Personality 299

by the party and germane to its functioning. These people
would not be expected to show identical attitudes with those
held by the major set of party members.

Our systematic analysis is relatively complete so far, but
we may still feel that a somewhat more detailed analysis of
the personality of members of the Communist and Fascist
parties might reveal more than is contained in the general
statement that they tend to be extraverted. This suspicion is
probably well-grounded, but unfortunately it is extremely
difficult to obtain the co-operation of members of extremist
groups for the purpose of studying their personality struc-
ture. Communists, on the whole, are rather more co-opera-
tive, but Fascists are extremely suspicious and distrustful,
and refuse point-blank any requests for co-operation. In
the circumstances, comparatively little has been done, but
even the few isolated results which are available are of con-
siderable interest.

One such study carried further the hypothesis, outlined
near the beginning of this chapter, that a special character-
istic of Fascist and Communist groups would be their aggres-
siveness, i.e. their failure to become properly conditioned to
the social prohibitions regarding the open expression of vio-
lence against other people. Forty-three Communists and
forty-three Fascists were studied in this particular experi-
ment, and their reactions compared with those of a group of
eighty-six people equated with them from the points of view
of age, class, and social status, but differing in that they held
political views favouring one of the three democratic parties.
In addition to being given the radicalism-conservatism ques-
tionnaire, these three groups were also tested by means of
the Thematic Apperception Test, described in an earlier
chapter. Special attention was paid in the analysis of the
stories the subjects told to evidence of aggression of both an
overt and a covert nature. Scoring along these lines is quite
reliable when the scorers adhere to the definition of aggres-
sion adopted for this purpose, which was 'To hate, fight, or
punish an offence. To criticize, blame, accuse, or ridicule
maliciously. To injure or kill, or behave cruelly. To fight

300

Part Two • Personality and Social Life

against legally constituted authorities; to pursue, catch or
imprison a criminal or enemy.5 Each person was given a
score according to the number of times that clear evidence
of aggression was found in his stories. The result of the ana-
lysis is shown in Figure 9, where scores on the radicalism-

• Fascists
+ Communists
fishMeanof
^^ Neutral Group

* + •

+ +

h 4- ++ + +

+ •

• a e «

S

.9

Radicalism

Conservatism

30

29

28

27

26

25

24

23

22

21

20

19

18

17

16

15

14

13

12

U

10

9

8

7

6

16 15 14 13 12 11 10 9 8 7 6 5 4 3 2

Figure 9 : Aggression scores of Fascists and Communists
as compared with a neutral group

conservatism scale are shown on the abscissa, and aggression
scores on the ordinate. It will be seen that each and every
one of the Communists had aggression scores which were in
excess of the mean of the neutral group, i.e. the group of
people voting for the three democratic parties. The same,
with very few exceptions, is true of the Fascists who took

Politics and Personality 30 1

part in this experiment; only four of these have aggression
scores slightly lower than the neutral group. All the others
have aggression scores very much higher than the neutral
group.

An inspection of the actual stories told by Fascists and
Communists reveals them to be dripping with blood.
This is particularly true of many of the stories told by
the Fascists; the amount of aggressiveness found in these
stories is quite beyond the range of what is found in normal
people. It will be seen from the Figure that scores as high as
thirty are not unusual among Fascists as compared with the
mean for the normal group of about eleven. Communists, by
contrast, are somewhat more aggressive than average, but
not abnormally so; their mean is in the neighbourhood of
sixteen. These data suggest certain differences between
Communists and Fascists, but for want of more detailed in-
vestigations it is not possible to follow up this lead.

Other investigations of a similar nature have shown Fas-
cists and Communists to be more dominant than members
of the democratic parties, to show a certain amount of rigid-
ity, and some intolerance of ambiguity. Suggestive as these
findings are, it must be obvious that much more thorough
experimental inquiries are needed before we can claim to
have gained but a superficial understanding of the personal-
ity dynamics which cause a person to become a member of
the Communist or Fascist party.

The reader, comfortably seated in his arm-chair in front
of the fire, and glancing through the pages describing these
results, will almost certainly feel, as does the writer, that they
do little more than whet the appetite, and he may wonder
why more has not been done in this field. One of the main
reasons is the great difficulties which lie in the path of the
investigator. Let us take but one example. To have adminis-
tered the Thematic Apperception Test and a few question-
naires to forty-three Fascists and forty-three Communists
may not sound a very considerable task. Yet the student who
did this work had to spend approximately a year in simply
gaining access to meetings held by these parties, obtaining

302 Part Two • Personality and Social Life

the confidence of a few members in each, and thus prepar-
ing the ground for the heart-breakingly difficult task of in-
dividually persuading forty-three members of each group to
undergo the testing programme. All this had to be done
without revealing the purpose of the experiment, without
losing the co-operation of a member once he had been ap-
proached (this would have upset the sampling procedure),
and without allowing the members of one party to suspect
that she was in any way friendly with members of the other
party. Almost every Saturday evening, come rain or shine,
snowstorm, blizzard, or hail, this intrepid young lady at-
tended open-air political meetings; most of her evenings
were spent arguing, debating, and reading party literature
so as to be able to talk the accepted jargon. Inevitably, per-
sonal complications arose which had to be resolved. All
through the time that this work was being carried on there
was a risk of personal danger if any suspicions had been
allowed to arise with respect to her exact role. Few students
are willing, or capable, of carrying out scientific research of
a high standard under these conditions, and the majority
rest content with the less interesting, but more easily obtain-
able types of data.

In spite of all these obvious and very great difficulties, it
might be possible to induce some of the more adventurous
students to take up research of this kind if society showed
some interest in the results so laboriously acquired, but un-
fortunately experimental social science is not welcomed very
much in academic quarters, where the quiet somnolence of
the reading-room and the dead and forgotten writings of
past nonentities are considered much more soothing than
the fresh air of empirical investigation, and the intoxicating
flood of factual data. Until this general attitude changes, it
is idle to expect any great access of knowledge in these com-
plex and difficult fields.

KEY TO SOCIAL ATTITUDE INVENTORY

The scoring key for the two scales is given after each of the items.
There are sixteen items for the measurement of R and thirty-two

Politics and Personality 303

items for the measurement of T; some items are used for measuring
both dimensions. Some items in the scale are 'filler' items and are
not scored at all. As regards scoring, the R scale is always scored in
the radical direction. For items marked R+ in the key, agreement
(+ or ++) is scored 1, and any other response o. For items

marked R— , disagreement (—or ) is scored 1, and any other

responses o.

The T scale is always scored in the tender-minded direction.
For items marked T + , agreement (+ or ++) is scored 1, and

Tough

6

7

a Fascists

9

10

11

12

Communists

®

R-

13 12 11

!0

®

Socialists

Conservatives

15
16

u® Liberate

18
19
20
21

22

Tender

Figure 10: Empirically determined positions of Communists, Socialists,

Liberals, Conservatives, and Fascists on two main dimensions

any other response o. For items marked T— , disagreement

( — or ) is scored 1 and any other response o. The range of

scores in the T scale is from o to 32 ; the range of scores in the R
scale is from o to 16.

In comparing these scores with those of members of various
political groups, the reader will find Figure 10 useful. In this are
shown the actual mean scores of Communists, Fascists, Socialists,
Liberals, and Conservatives on the R and T scales. By entering
his own score in this table, the reader will be able to see how he
stands with respect to the major political organizations in this
country.

304 Part Two • Personality and Social Life

Below are given sixty statements which represent widely-held opinions
on various social questions, selected from speeches, books, newspapers,
and other sources. They were chosen in such a way that most people are
likely to agree with some, and to disagree with others.

After each statement, you are requested to record your personal
opinion regarding it. You should use the following system of marking:

+ + if you strongly agree with the statement

+ if you agree on the whole

o if you can't decide for or against, or if you think the question is

worded in such a way that you can't give an answer
— if you disagree on the whole
if you strongly disagree

Please answer frankly. Remember this is not a test; there are no 'right'
or 'wrong' answers. The answer required is your own personal opinion.
Be sure not to omit any questions. The questionnaire is anonymous, so
please do not sign your name.

Do not consult any other person while you are giving your answers.

Opinion Statements Tour Opinion

1 . The nation exists for the benefit of the individuals
composing it, not the individuals for the benefit of
the nation.

2. Coloured people are innately inferior to white
people.

3. War is inherent in human nature.

4. Ultimately, private property should be abolished

and complete Socialism introduced. R-h

5. Persons with serious hereditary defects and diseases
should be compulsorily sterilized.

6. In the interests of peace, we must give up part of our
national sovereignty.

7. Production and trade should be free from govern-
ment interference. R —

8. Divorce laws should be altered to make divorce

easier. T —

9. The so-called underdog deserves little sympathy or

help from successful people. T—

10. Crimes of violence should be punished by flogging. R— T—

11. The nationalization of the great industries is likely

to lead to inefficiency, bureaucracy, and stagnation. R—

12. Men and women have the right to find out whether
they are sexually suited before marriage (e.g. by trial

marriage). R+ T—

13. 'My country right or wrong' is a saying which ex-
presses a fundamentally desirable attitude. R—

Politics and Personality 305

Opinion Statements — contd Tour Opinion

14. The average man can live a good enough life with-
out religion. T —

15. It would be a mistake to have coloured people as
foremen over whites.

16. People should realize that their greatest obligation
is to their family.

17. There is no survival of any kind after death. T —

18. The death penalty is barbaric, and should be abol-
ished. R+ T+

19. There may be a few exceptions, but in general, Jews

are pretty much alike. T —

20. The dropping of the first atom bomb on a Japanese
city, killing thousands of innocent women and chil-
dren, was morally wrong and incompatible with our

kind of civilization. T-f-

2 1 . Birth control, except when recommended by a doc-
tor, should be made illegal. T-f-

22. People suffering from incurable diseases should have

the choice of being put painlessly to death. T —

23. Sunday-observance is old-fashioned, and should
cease to govern our behaviour.

24. Capitalism is immoral because it exploits the worker
by failing to give him full value for his productive
labour. R -f-

25. We should believe without question all that we are
taught by the Church. R —

26. A person should be free to take his own life, if he
wishes to do so, without any interference from

society. T —

27. Free love between men and women should be en-
couraged as a means towards mental and physical

health. R+ T-

28. Compulsory military training in peace-time is essen-
tial for the survival of this country. T —

29. Sex crimes such as rape and attacks on children, de-
serve more than mere imprisonment; such criminals
ought to be flogged or worse. R —

30. A white lie is often a good thing. T —

31. The idea of God is an invention of the human mind. T —

32. It is wrong that men should be permitted greater
sexual freedom than women by society.

33. The Church should attempt to increase its influence

on the life of the nation. T4-

34. Conscientious objectors are traitors to their country,
and should be treated accordingly.

306 Part Two • Personality and Social Life

Opinion Statements - contd Tour Opinion

35. The laws against abortion should be abolished.

36. Most religious people are hypocrites. T —

37. Sex relations except in marriage are always wrong. R— T-j-

38. European refugees should be left to fend for them-
selves. T—

39. Only by going back to religion can civilization hope
to survive.

40. It is wrong to punish a man if he helps another coun-
try because he prefers it to his own. R-f-

41 . It is just as well that the struggle of life tends to

weed out those who cannot stand the pace. T —

42. In taking part in any form of world organization,
this country should make certain that none of its
independence and power is lost. R —

43. Nowadays, more and more people are prying into

matters which do not concern them. T —

44. All forms of discrimination against the coloured
races, the Jews, etc., should be made illegal, and
subject to heavy penalties.

45. It is right and proper that religious education in
schools should be compulsory.

46. Jews are as valuable citizens as any other group. T+

47. Our treatment of criminals is too harsh; we should

try to cure them, not punish them. R-f- T-f

48. The Church is the main bulwark opposing the evil

trends in modern society. T+

49. There is no harm in travelling occasionally without

a ticket, if you can get away with it. T—

50. The Japanese are by nature a cruel people.

51. Life is so short that a man is justified in enjoying

himself as much as he can. T —

52. An occupation by a foreign power is better than war. R-f- T+

53. Christ was divine, wholly or partly in a sense differ-
ent from other men. T-f-

54. It would be best to keep coloured people in their
own districts and schools, in order to prevent too
much contact with whites.

55. Homosexuals are hardly better than criminals, and
ought to be severely punished.

56. The universe was created by God. T-f-

57. Blood sports - like fox-hunting, for instance - are

vicious and cruel, and should be forbidden. T-f-

58. The maintenance of internal order within the nation
is more important than ensuring that there is com-
plete freedom for all. T—

Politics and Personality 307

Opinion Statements - contd Tour Opinion

59. Every person should have complete faith in some
supernatural power whose decisions he obeys with-
out question.

60. The practical man is of more use to society than
the thinker.

8

THE PSYCHOLOGY OF AESTHETICS

There can be few topics more certain to lead to furious
discussion than those related to aesthetics; there can be few
topics within the realm of aesthetics more certain to arouse
normally peaceful artists, philosophers, and aestheticians to
a pitch of uncontrolled indignation than that which has
given this chapter its title. The idea that objects of beauty,
as well as their creation and appreciation, are subject to
scientific scrutiny appears abhorrent to most people, even as
the idea that the physicist might study and analyse the
colours of the rainbow with his objective methods was ab-
horrent to their grandparents. There appears to exist a fear
that clumsy handling might crush the butterfly's wings; an
idea that analysis may destroy what it is intending to study.

Associated with this fear is perhaps another. Most people
hold views regarding aesthetics which they are extremely
unwilling to give up, although these views are not based on
any objective facts. Indeed, the very idea that one's views
ought to be related to factual evidence is usually dismissed,
and it is asserted that subjectivity reigns supreme in this
field. This, of course, is a tenable view; it is contradicted,
however, by the well-known tendency of most people to
argue about their aesthetic views, often with great acerbity,
always with great tenacity, never with that humility which
the hypothesis of complete subjectivity should engender in
them. If aesthetic judgements are completely subjective,
there would appear as little point in argument as in scien-
tific experiment; if the one is permissible, so surely is the
other. Perhaps the objection to scientific investigation is in
part due to a fear that facts may be more potent than argu-
ments in forcing one to give up a cherished position, and to
acknowledge certain objective factors which one would pre-
fer to overlook.

However that may be, there can be little doubt about the

308

The Psychology of Aesthetics 309

hostile reaction which psychology has experienced on all
sides when it attempted to introduce scientific methods into
the study of aesthetics. A good deal of this hostility is prob-
ably based on misunderstanding, and it will be the purpose
of this chapter to explain in some detail just what the psycho-
logist is trying to do, and how he sets about his task. I shall
try to avoid arguments and comparisons with philosophical
procedures and problems as far as possible ; these often seem
to resemble those attacked by psychologists, but the simi-
larity is only superficial. The reader familiar with modern
aesthetic doctrines, and with the long history of discussions
in this field, will easily be able to apply the facts of psycho-
logical research to the solution of such philosophical prob-
lems as interest him.

How, then, does the psychologist start? He notices that
certain types of judgement are made frequently of certain
objects; these judgements are phrased in terms of 'beautiful'
and 'ugly', or some synonymous terms, and apply to various
combinations of colours and shapes, as in the visual arts;
words, as in poetry; or sounds, as in music. The essential
datum with which he deals, therefore, is a relation - a rela-
tion between a stimulus (picture, poem, piece of music) and
a person who reacts to this stimulus in certain conventional
ways. Usually the response is a verbal one, but it is possible,
and has been found useful in certain situations, to record
physiological reactions indicative of emotion, such as heart-
beat, pulse-rate, skin temperature, or changes in the electric
conductivity of the skin.

In analysing this relation, the psychologist encounters a
twofold problem. In the first place, he must ask himself:
Just what is the physical property of the stimulus which
causes a favourable reaction as opposed to an unfavour-
able reaction in the majority of the subjects with whom
he is working? In the second place, he must ask himself:
Just what is the reason why one person reacts favour-
ably to a given stimulus, while another person reacts un-
favourably? Possible answers to the first question might be
in terms of certain 'laws of composition', as for instance that

310 Part Two • Personality and Social Life

the centre of interest of a picture ought to lie at the inter-
section of the 'thirds', i.e. of lines drawn parallel to the side
and bottom of the pictures, and dividing it into three equal
parts either way. Answers to the second question might be
in terms of temperamental traits or types; thus it might be
argued that introverts prefer classical, extraverts romantic
music. I am not here arguing that these examples in any way
correspond to fact; they are merely quoted to show possible
methods of answering psychological questions.

Inevitably, the psychologist will start his work by experi-
menting with the simplest possible stimuli - simple colours
and colour combinations, simple proportions of lines, and so
forth. In doing this he is following the usual path of scientific
progress from the simple to the complex. It is here that he
frequently encounters the first serious objection on the part
of the philosopher and aesthetician, who claims that judge-
ments regarding the relative beauty of simple colours or lines
are not in any way related to judgements of more complex
stimuli, such as a landscape by Cezanne or a portrait by Rem-
brandt, and that consequently rules and laws derived from
simple stimuli can have no relevance to what are considered
'real' works of art. No proof is offered for this rejection of
evidence, other than the subjective feeling of the critic that
these judgements are 'qualitatively' different. I shall not
attempt to argue this point here, but shall defer discussion
until later on, where evidence will be cited to prove quite
definitely that there are essential similarities linking aes-
thetic judgements of 'simple' with aesthetic judgements of
'complex' stimuli.

How does the psychologist design his experiment? Usually
he will provide a series of stimuli whose physical properties
are known, and ask his subjects to rank these in order of aes-
thetic merit, i.e. from best liked to least liked. Alternatively
he may offer his subjects two stimuli at a time, with the re-
quest to say which of the two is more pleasing aesthetically;
all possible combinations of stimuli are shown in this man-
ner. Either procedure will result in an average order of pre-
ference, and experience has shown that this order will be

The Psychology of Aesthetics 311

pretty much the same regardless of the exact method used
for deriving it. From this average order of 'aesthetic' merit,
certain deductions may be made regarding the physical pro-
perties associated with high-ranking and low-ranking
objects respectively.

It is here that a second objection will often be made.
Psychologists, it will be said, are treating the perception of
'beauty' as if it were essentially similar to some 'objective'
property like greenness, or size, or shape. But this can hardly
be permissible; surely 'beauty' is not a property belonging
to an object in the same way that one might say the colour
green, or the triangular shape, belonged to the object.
Beauty, in other words, is essentially subjective; colour,
shape, and other properties of a stimulus are objective. How
can one reasonably use methods appropriate to the study of
one type of stimuli in the study of other, different types?

This objection is based on an essential fallacy, a fallacy
whose hoary age does not prevent it from coming up again
and again in modern discussion. An object does not 'con-
tain' the colour green in any meaningful sense of the term;
it reflects light of a certain wavelength which some people
experience as 'green', others, who happen to be colour-
blind, as 'grey'. Similarly, an object does not 'contain'
beauty in any meaningful sense of the term ; it reflects light
in certain combinations of wavelengths which some people
experience as 'beautiful', others as 'ugly' or 'indifferent'.

Some people, harking back to Locke's distinction between
primary and secondary qualities, are willing to concede the
force of this argument with respect to colour, but balk at its
extension to shape. Here, they would say, there is complete
correspondence between stimulus and experience ; everyone
sees a circle as round, a triangle as different from a square.
Alas, the facts contradict even this confident assertion. Ex-
periments with people whose congenital blindness was re-
moved surgically in later life, and who thus experienced
sight for the first time, have shown them to be quite incap-
able of distinguishing between a circle and a square, or of
recognizing triangles and other simple figures. Weary

312 Part Two • Personality and Social Life

months oflearning were needed for them to make even such
very simple discriminations, and the disheartening slowness
with which such learning proceeded bore ample testimony
to the absurdity of the notion that ' roundness ' or ' square-
ness' were inherent qualities in the object, just waiting to be
perceived. Rules for the perception of these qualities had
to be acquired, just as we have to acquire rules for the per-
ception of beauty; without these rules there is literally no
perception at all.

This fact was brought out with particular clarity in ex-
periments on animals, mainly chimpanzees and rats, where
the animal was reared in darkness. Although there was no
interference whatever with the physiological apparatus of
vision, the animals, when brought into the light, behaved to
all intents and purposes as if they were blind; they could not
learn to avoid a large, distinctive object from which they
obtained strong electric shocks, and they failed to learn to
recognize the white-clad attendant to whom they were at-
tached, in spite of the fact that he stood out conspicuously
from the uniform grey background. Perception of colours,
shapes, and other physical properties is a learned activity,
and what is perceived depends very strongly on the type of
learning and the amount of learning which the animal - or
the human - has gone through. In this respect the percep-
tion of 'beauty', therefore, is no different from the percep-
tion of other qualities.

These facts make it necessary to inquire into the precise
meaning of those terms, 'objective' and 'subjective', which
are so often used to mark the distinction between properties
which are supposed to be capable of being investigated by
scientific techniques - such as colour and shape - and others
which are not - such as beauty. 'Objective' is usually taken
as synonymous with 'real', 'subjective' with 'unreal'. But
we have shown that to call a stimulus 'green' is far from
being an 'objective' description; all that we can say objec-
tively is that the stimulus object reflects light-waves of a cer-
tain periodicity. The experience 'green' is subjective, i.e. in-
herent in the observer rather than a characteristic of the

The Psychology of Aesthetics 313

stimulus. If it is nevertheless permissible to try to link up the
subjective experience and the objective stimulus in the case
of colour or form perception, it is difficult to see why it
should not be permissible to do the same in the case of our
perception of 'beauty'.

Here the argument often changes its content, and the
term 'objective' assumes a different meaning. It is said that
everyone is agreed on the experience of 'green' when his
eyes are stimulated by a light of the wavelength of 515 milli-
microns, while he will report an experience of 'red' when
the wavelength changes to 650 millimicrons. But there is no
such agreement with respect to experiences of 'beautiful'
and 'ugly'; one man's meat is another man's poison, and
de gastibus non est disputandum. In other words, 'objectivity'
is now defined in terms of agreement among observers;
where such agreement obtains, as it does in the case of
colour judgements among individuals with perfect colour
vision, the judgement is said to be objectively based. Where
there is no agreement, judgement is said to be subjective. We
may accept this type of definition, but we should be aware
that in so doing we abandon the absolute distinction be-
tween 'objective' and 'subjective', and recognize instead
degrees of 'objectivity' depending on the amount of agree-
ment observed among our subjects. In other words, our de-
cision as to the 'objectivity' of a judgement ceases to be
determined by philosophical argument, and becomes in-
stead an empirical and experimental question, to be settled
by observations regarding the degree of agreement found.
This is the sense in which the term will be used here.

As an example of experimental work in this field, let us
take the numerous studies on colour preferences. In order to
appreciate their outcome, we must first of all be able to
specify exactly the colour stimulus; unless we can do this,
we cannot describe our experiment completely, thus en-
abling others to repeat it, and we cannot state any laws re-
lating preference judgements to the physical properties of
the stimulus. Now essentially there are three dimensions
along which colours can differ from each other (black, white,

3H

Part Two • Personality and Social Life

and grey are called 'colours' here in addition to red, green,
yellow, blue, and the other chromatic colours) . These three
dimensions are known as hue, saturation, and brightness,
and their relationship is shown in the accompanying dia-
gram. Hue refers to the chromatic quality which distin-
guishes red from yellow, or blue from green; it is measured

white

Hue

Hue

Black

Figure 1 1 : Showing the three-dimensional
nature of colours

in terms of wavelength. Brightness refers to the amount of
light reflected by the colour, while saturation refers to the
amount of chromatic colour shown (its vividness) . Hues are
arranged in a circle (the so-called 'colour-circle') in such a
way that colours at opposite poles of the circle (comple-
mentary colours) give grey when mixed together.

If we take coloured chips representing the various parts of

The Psychology of Aesthetics 315

the colour circle, being careful to have them all of the same
brightness and saturation, and ask a number of people to
rank them in order of preference by means of either of the
two methods described earlier, we find that there is con-
siderable agreement between different people. This agree-
ment is still manifest when we are less careful to keep bright-
ness and saturation equal for all our colours, but it is much
less strongly marked, because judgements are now based not
on one characteristic (hue) only, but on a combination of
several. Nevertheless, considerable agreement is still found,
even when we compare European subjects with savage
tribes, American whites with Red Indians, and subjects from
Oriental with subjects from Occidental cultures. There thus
appears some strong biological foundation for judgements of
colour preference, a foundation which may occasionally be
overborne by cultural influences but which asserts itself in
all the many diverse groups studied.

There seems to be a definite physical property in the
stimulus which is responsible for this universal order of pre-
ference. Short wavelengths are generally preferred to long
wavelengths; the correlation between wavelength and pre-
ference is almost perfect. For young children this relation-
ship does not appear to hold, but for adolescents and adults
it appears to constitute a natural law.

If people differ in their preference judgements - and
agreement is far from perfect - then it would follow that
some people's judgements are more in accord with the aver-
age order of colours than are the judgements of other people.
If, in accordance with our definition of the term 'objective',
we call this average order of colour preferences the ' objec-
tive' or 'true' order, then we can perhaps call those who
agree with it most the 'best' judges, and those who agree
with it least the 'poorest' judges. Alternatively, we may say
that our 'best' judges have good taste, while our 'poorest'
judges have had bad taste; we would, in that case, be defin-
ing the terms 'good' and 'bad' taste in a somewhat unusual
way, and the reader is of course free to reject that definition
of the term. Evidence will be brought forward later on to

3 1 6 Part Two • Personality and Social Life

show that this suggested usage agrees quite well with com-
mon usage in many ways; for the moment let us just note
this new way of defining the concept.

We have dealt with preference judgements of hues of a
given saturation and brightness ; what happens if we give
the same subjects a test involving hues at different levels of
saturation and brightness? What would happen if we gave
them tests in which preference judgements were to be made
between colours of different degrees of saturation, keeping
hue and brightness constant, or between colours of different
degrees of brightness, keeping hue and saturation constant?
The outcome of these experiments is in line with the 'hue'
experiment already described; there is substantial agree-
ment between different judges, and those people who are
■ good judges ' (have good taste) in one test are on the whole
those who are ' good judges ' (have good taste) in the other
tests. The quality we have denoted as 'good taste' thus is not
dependent on any one particular test; it applies generally to
all the tests which can be constructed in the field of colour
preference.

What happens if we extend our work to colour combina-
tions - say combinations of two colours of equal brightness
and saturation, to keep the problem at a manageable level?
The answer to this question is important, for two reasons. In
the first place, aestheticians often maintain that judgements
regarding single colours are not aesthetic judgements at all ;
it is at the level of complexity represented by colour com-
binations that the simplest form of aesthetic judgement be-
gins. Thus a demonstration that what is true of simple
colour judgements is also true of judgements regarding
colour combinations is important in showing that the aes-
thetician's argument is possibly wrong, and that we may
generalize from simple colour experiments to more complex
stimuli.

Even more important is another argument. There is an
important school in psychology, the holistic or 'Gestalt'
school, which maintains that complex units or 'gestalten'
are not built up atomistically from simpler units or 'atoms' ;

The Psychology of Aesthetics 317

rather, the more complex unit shows 'emergent' qualities
which cannot be predicted from knowledge of the simpler
constituents and the relations obtaining between them. If
this argument were true, then our whole attempt to derive
laws governing appreciation of complex works of art from
experiments dealing with relatively simple objects would be
doomed to failure. Here we have an ideal testing ground for
the 'atomistic' hypothesis. If we can predict preferences for
colour combinations on the basis of knowledge of prefer-
ences for single colours, and knowledge of the relation on the
colour circle between the colours in each combination, then
we would have disproved the ' Gestalt' argument, and might
with reasonable assurance go on with our general plan. If
such prediction should prove impossible, then we would
have to abandon our ' atomistic' approach, and look around
for a different methodology.

First of all, let us note that with respect to colour com-
binations we again find a certain marked degree of agreement
or 'objectivity'. Secondly, let us note that again those who
prove to be 'good' judges on one test involving colour com-
binations also turn out to be 'good' judges on other tests in-
volving colour combinations. Thirdly, let us note that these
'good' judges of colour combinations are precisely those
who earlier on were found to be good judges of single
colours and their aesthetic values. Whatever constitutes
'good taste' in the one experiment obviously constitutes
'good taste' in the other; we can justifiably generalize from
simple to more complex stimuli.

But even more important than this is another demonstra-
tion. It is possible to show that preference judgements of
colour combinations depend on two factors. The first is the
simple sum of the preferences for the individual colours ; if
both the individual colours making up the combination are
liked, then the combination will on the whole be liked. If
both colours are disliked, the combination will tend to be
disliked. If one colour is liked, the other disliked, or if both
are neutral, then the affective value of the combination will
tend to be neutral.

3 1 8 Part Two • Personality and Social Life

The second factor relates to the position of the two com-
ponent colours on the colour circle. The closer together the
two colours are on that circle, the lower will be the aesthetic
ranking of the combination ; the further apart they are, the
higher will be the ranking of the combination. Best liked of
all are pairs of complementary colours, i.e. colours exactly
opposite each other on the colour circle.

If we combine these two factors - liking of the individual
colours, and knowledge of their separation on the colour
circle - then we can predict with very great accuracy indeed
the aesthetic ranking of the colour combination. Thus the
evidence decisively favours the atomistic hypothesis, and
does not agree at all with the holistic view that in the appre-
ciation ol complex objects 'emergent' qualities come into
play which cannot be dealt with in terms of simple qualities
and their relations. This conclusion may of course have to be
modified when dealing with objects of very high complexity,
like landscape paintings or portraits, but even so this finding
is certainly encouraging.

So far we have been dealing exclusively with colours, and
the question must be faced : can we extend our findings to
other properties of works of art? One such set of properties
is dealt with in the 'laws of composition' referred to already,
and it seems worth while to inquire into the possible rela-
tion between 'good taste' as defined in our experiments so
far, and 'good taste' as it might be defined by these laws.

Fortunately there exist tests constructed for the express
purpose of obtaining a measure of this ability to judge the
factors entering into good composition. This is usually done
by contrasting two line drawings, or two designs, one of
which purposely violates one of these rules, while the other
is unimpeachable on this account. Great care is taken in the
construction of these tests to obtain the best advice avail-
able from artists, art teachers, and art critics, and only if
these experts are practically unanimous as to the respective
aesthetic value of the two designs making up each test item
are they included in the test. Here, then, we have rather a
different criterion of 'good taste', one much nearer to the

The Psychology of Aesthetics 3 1 9

way in which the term is used in ordinary speech. It would
of course still be possible to argue that the unanimous ver-
dict of all the experts who had devoted their lives to the
practice and study of painting was mistaken, and that their
standards were quite arbitrary ; such a nihilistic view would
find it very difficult indeed to account for some of the find-
ings to be reported presently.

In the first place, tests of this type predict with consider-
able accuracy which students in the arts school make a suc-
cess of their studies, and which fail ignominiously. The tests,
be it noted, do not predict whether the person with a high
score is likely to paint in the classical manner, or whether he
will become a revolutionary; they merely predict that he is
likely to paint well regardless of the particular manner he
chooses, or the style which he finally adopts. The contro-
versy over 'modern' painting has blinded many people to
the fact that paintings differ in quality as well as in style, and
that one can paint well or badly in any style; it is this quality
which tests attempt to predict, and the evidence shows that
they are successful in doing so at least to some extent.

In the second place, and this point cannot be stressed too
strongly, it has been found that people who show good taste
in their judgements of simple colours and of colour com-
binations also do well on these completely achromatic tests
of composition. This finding must certainly be somewhat un-
expected to the subjectivist; it is accountable in terms of the
hypothesis that there exists some property of the central ner-
vous system which determines aesthetic judgements, a pro-
perty which is biologically derived, and which covers the
whole field of visual art. People would on this hypothesis be
expected to differ with respect to 'good taste' in the same
way in which they are known to differ with respect to acuity
of vision, ranging from an extreme of Philistine lack of all
aesthetic appreciation - a true ' blindness' to all that is beau-
tiful - to the other extreme of almost instinctive appreciation
of the good and beautiful, and abhorrence of the bad and
ugly. Admittedly, this is only a hypothesis at this stage, but
it appears to be the only hypothesis to account for all the

320 Part Two • Personality and Social Life

facts, and it is scientifically valuable in that verification or
disproof can be very easily arranged. One deduction, for in-
stance, might be that this ability should be very strongly
determined by heredity; there is already some evidence in
favour of this view, but it would require experiments with
identical and fraternal twins to make the proof conclusive.

Another deduction might be that a person who showed
good taste (as defined) with respect to one type of visual art
should also show good taste with respect to any other type
of visual art. This deduction has been verified by construct-
ing tests involving a great variety of different types of visual
stimuli - portraits, landscape paintings, book-bindings,
silverware, statues, landscape photographs, carpets, and
many more. In each case it was found that the person show-
ing good taste on one test tended also to show good taste on
the others, just as predicted by our hypothesis. This finding
also is very difficult to account for in terms of any alternative
hypothesis, although several such may have suggested them-
selves to the reader.

In the first place, it might be argued, this agreement may
be due entirely to intelligence; the more intelligent may also
be the more 'artistic', and perhaps the more knowledgeable
regarding aesthetic values. This hypothesis falls to the
ground because intelligence correlates only to a very slight
extent with 'good taste'; certainly the correlation is much
too low to account for the findings.

A more reasonable hypothesis might be one which re-
ferred the observed correlations to cultural factors entirely;
the argument might run something like this. A person who
is knowledgeable with respect to current views about the
aesthetic value of certain paintings would also be know-
ledgeable with respect to current views about the aesthetic
value of different types of carpets, or statues, or book-bind-
ings. Thus the tests might merely measure 'cultural know-
ledge', rather than something more fundamental.

This argument also cannot be defended. The specimens
used as test objects were all unknown to the subjects taking
part in the experiment; they had been selected with particu-

The Psychology of Aesthetics 321

lar care in order to obviate any criticism of the experiment
along the lines of the 'cultural' hypothesis. When pictures
were taken from the works of an artist relatively well known,
then the whole test was made up of pictures taken just from
that one artist, so that there was no 'cultural influence' or
'general attitude' determining the relative value of the
different items. Even more conclusive is the last considera-
tion; how could cultural influences create a correlation be-
tween aesthetic tests of the type described, and simple colour
judgements? There is no cultural ruling which some subjects
might know, and with which they might conform, in the
field of colour rankings. Indeed, most of the subjects doubt-
ed if there was any agreement at all among people in this
field, holding strongly to a 'subjectivist' position. But an in-
fluence which is absent in the case of one type of test cannot
possibly cause a correlation between that test and others;
consequently we must dismiss the 'cultural' argument, in
spite of its superficial appeal.

This point should not be stretched too far; it certainly is
not denied that cultural influences have very great import-
ance indeed. When we look at the factors which determine
the judgements of many people in the field of art, we find
that some of these factors are not of an aesthetic nature at all.
The monetary value of the picture, its fame, the fact that it
is exhibited by the Royal Academy, knowledge of the posi-
tion of the painter in the hierarchy of his colleagues - all
these and many other extraneous considerations determine
what people will say when asked : ' Do you like this picture? '
But the psychologist - no more than the aesthetician and the
philosopher - is not particularly interested in these irre-
levant factors; he wishes to isolate the determinants of
genuinely aesthetic responses. In order to do that he has to
select his material carefully, so that considerations of the
type described cannot influence his subjects. Such control of
irrelevant factors is absolutely essential ; without it we would
be lost in a welter of contradictory and non-aesthetic deter-
minants.

Similar arguments apply to another troublesome question.

322 Part Two • Personality and Social Life

If we make up a test consisting not only of pictures of
widely different quality, but also contrasting strongly in
style - including, say, Old Masters and works of the Picasso-
Matisse school - then style preferences would to some extent
obscure the differences in quality which are the main object
of our interest. Consequently, such a test would not be a
good test for our purpose; we would have to split it into
two, making up one test exclusively of Old Masters, the
other exclusively of works of the Picasso-Matisse school. In
this way we would prevent style differences from interfering
with our measurement of differences in our subjects' ability
to judge the quality of the pictures.

Can we justifiably neglect such important factors as those
associated with the Ancient vs. Modern controversy? The
answer is surely that we cannot neglect them in any orderly
description of the whole of aesthetic appreciation, but that
we must pass them by in our attempt to isolate and measure
one particular aspect of aesthetics, namely the qualitative
one, which appears to be largely independent of the contro-
versy. We cannot give an adequate description of a person's
body by measuring his height alone, but when we are mea-
suring his height we disregard for that purpose such factors
as his weight, his flat feet, and the wart on his nose. For cer-
tain purposes measures of height are valuable, for others
they are almost useless; what we want to measure depends
upon our purpose. If quality and style of painting are inde-
pendent variables - and the evidence in favour of this view
is very strong - then we must measure them independently
and in isolation. If there are still other factors, then they also
must be investigated and measured, each in its turn.

In actual fact, measurement of style preferences is very
much easier than measurement of quality judgements. The
usual method has been to select sets of two pictures, both of
which depict a similar scene, a windmill, say, or a waterfall ;
one of these is taken from the paintings of a well-known
modern painter, the other from the paintings of an equally
well-known classical painter. In this way we may hope to
keep under control interest in the subject-matter, the quality

The Psychology of Aesthetics 323

of the painting, and the acceptability of the artist's name;
preference judgements as between the two paintings should
then be strictly a measure of 'style' preferences. Studies
along these lines have shown fairly convincingly that these
preferences are related to temperament; introverts tend to
prefer the older, extraverts the more modern works. Inter-
esting as this finding is, we cannot stop to discuss it here, as
it is somewhat outside the main argument.

To return, then, to our muttons, we may be said, in a
way, to have taken the average order of preference of the
population as our standard of 'good taste'. This motion is so
alien to the most cherished tenets of aestheticians and philo-
sophers that it is liable to be ridiculed on irrelevant grounds.
Thus it might be said that surely Mozart is a greater com-
poser than Irving Berlin, in spite of the fact that the great
majority seem to prefer the latter's productions. A Holly-
wood musical is aesthetically inferior to the Midsummer
Night's Dream, however many more customers may be at-
tracted to the former. How, then, can we use the average
judgement of the masses as the arbiter of aesthetic excel-
lence?

This criticism misses the whole point of the argument.
The average rank order of works of art is a good criterion of
excellence only under carefully specified conditions', all irre-
levant and extraneous factors must first be ruled out before
we can accept the average judgement as having any value
at all. Preference for a 'leg show' or the more 'bosomy*
Hollywood productions over Shakespeare may be a good
measure of strength of sexual interest; it is irrelevant to the
respective aesthetic values of these two productions because
the judgement is not based on aesthetic grounds. The useful-
ness and value of an average in science depend entirely on
the question asked, the conditions of the experiment, and
the precise nature of the figures averaged; under appro-
priate conditions, such an average may be of the greatest
value, under other conditions it may be useless and mis-
leading. Criticisms of the notion that the average ranking of
aesthetic objects can furnish us with an acceptable criterion

324 Part Two • Personality and Social Life

of aesthetic value are usually based on examples in which all
the rules for obtaining a meaningful average are broken;
this may make for a good knock-about argument, but it does
not help much in the search for scientific criteria of ' beauty \

One last point should be considered. In addition to
general 'good taste' and style preferences, aesthetic judge-
ments are often determined by highly individualistic and
idiosyncratic factors. A man may like yellow because his girl
always wears yellow; or he may show an inordinate prefer-
ence for Pechstein's ' Lupowmiindung ' because this picture
reminds him of sunny summer holidays on the sandy beaches
of the Baltic. These are extraneous factors which may be of
interest in themselves, but which do not affect the determina-
tion of our average order; being specific to one individual,
they tend to cancel out over large numbers. Essentially, this
type of preference determinant is non-aesthetic in nature,
being mainly based on associations with particular events
which have brought happiness or pain to the individual
concerned.

I have no space here to deal with the considerable body of
experimental work that has been done in the fields of music
and poetry; by and large, findings there are similar to those
in the visual arts. Nor is it possible to discuss aesthetic crea-
tion; too little is known in that field, apart from some in-
genious but unlikely Freudian speculations, to make any
reasonable conclusions possible. On the whole, we may say
that experimental work in aesthetics has unearthed a num-
ber of facts which cannot be disregarded by anyone inter-
ested in the problem of the formation of aesthetic judge-
ments, and that these facts all point with remarkable un-
animity to a theory of aesthetics which is firmly anchored in
biology and derives judgements of 'beauty5 from inherited
properties of the central nervous system. Over-simplified
and inadequate to deal with the tremendous complexities of
great works of art? Certainly. But nevertheless a beginning
which promises to lead us to a more adequate account in
due course.

It will be noted that our concern has been almost entirely

The Psychology of Aesthetics 325

with what are called formal elements of art. Would it be
possible to go a step further than we have done so far and
actually try to describe a formula, of a completely objective
nature, for the measurement of beauty? The attempt to do
this is not essentially modern in origin; it will be remem-
bered that Pythagoras attempted to base the beauty of
music on mathematical relations obtaining between the
lengths of the strings used in striking a chord, and other
writers of antiquity have at times made similar attempts.
Thus, Plato has this to say in the Timaeus: 'That triangle
which we consider to be the most beautiful of all ... is that
of which the double forms a third equilateral triangle.' His
reason, as will be seen from Figure 12, was that out of this
triangle could be built the equilateral triangle, the rect-
angle, the parallelogram, the diamond, and regular hexagon

AAttA?

Figure 12: Diagram showing the multiplicity
of relationships which caused Plato to call
the triangle on the extreme left 'the most
beautiful of all '

among polygonal figures, as well as three of the five regular
solids. This power in combination was peculiarly significant
to Plato, who valued it for purposes of cosmological specula-
tion.

We would hardly be content to agree with this or regard
his reasons as being of an aesthetic nature. This capacity for
missing the point has, of course, always been characteristic
of philosophers. (One of the extreme examples of this is
Schopenhauer's well-known essay on humour. He spends
some sixty pages in an attempt to analyse the particular ex-
cellence of what he regards as the best joke he has ever come
across. This joke, apparently - and it is here that there is an
interesting parallel with Plato - consists of a tangent to a
circle. The exquisite humour of the line approaching the
circle and then receding from it again provokes rhapsodies

326 Part Two • Personality and Social Life

of appreciation from Schopenhauer, which, however, have
not been echoed by most readers!)

It would be boring to review the whole history of attempts
of this kind, but we must mention one psychologist who may
be said to have elevated the study of aesthetics into a scien-
tific discipline. Fechner was particularly interested in the
experimental determination of preferences for proportions,
and tried to relate these to a well-known aesthetic doctrine;
namely, that of the golden section. This section of a linear
segment is that which divides it into two segments in such a
way that the longer segment is the mean proportional be-
tween the shorter segment and the whole segment. Fechner's
particular interest concentrated in the so-called * golden
rectangle', i.e. a rectangle whose sides are in the ratio of the
golden section. These rectangles, with the ratio of the longer
to the shorter side of 1-618, or very nearly 8 to 5, were sup-
posed to have some occult beauty by philosophers and
aestheticians, which made them quite outstandingly superior
to other types of rectangles.

Experimental work by Fechner, and many of his succes-
sors, has shown that rectangles having proportions some-
what similar to the 'golden rectangle' are indeed well liked.
It has also been found, however, that the exact proportion
of the sides required by the alleged law is not conspicuously
superior to neighbouring ratios, and, in fact, it is often
found to be inferior to them. Thus, there appears nothing
very occult or mystic about this ratio, and the general
theory endowing it with special beauty must remain very
suspect.

In recent years, an American mathematician, George D.
Birkhoff, has taken up this problem of an aesthetic formula
again in his book on 'Aesthetic Measure'. In this he has
made a thoroughgoing attempt to provide a general formula
for the measurement of works of art (visual art, poetry,
music). Three elements enter into this formula and are re-
lated to the three phases in the aesthetic experiences which
he distinguishes: (1) a preliminary effort of attention, which
is necessary for the act of perception, and which increases in

The Psychology of Aesthetics 327

proportion to what he calls the complexity (C) of the object;
(2) the feeling of value or aesthetic measure (M) which rewards
this effort; and finally (3) a realization that the object is
characterized by a certain harmony, symmetry, or order (O),
more or less concealed, which seems necessary to the aes-
thetic effect. Taking his cue from the well-known aesthetic
demand for 'unity in variety', he maintains that this ana-
lysis of the aesthetic experience suggests that the aesthetic
feelings arise primarily because of an unusual degree of har-
monious interrelation within the object. He goes on to say,
1 More definitely, if we regard M, 0, and C as measurable
variables, we are led to write

M=c

and thus to embody in a basic formula the conjecture that
the aesthetic measure is determined by the density of order
relations in the aesthetic object. . . . If we admit the validity
of such a formula, the following mathematical formulation of
the fundamental aesthetic problem may be made: Within
each class of aesthetic objects, to define the order 0 and the complexity
C so that their ratio M= OJC yields the aesthetic measure of any
object of the class*

Much of Birkhoff 's work is concerned with defining pre-
cisely how this order in complexity elements may be mea-
sured. In the case of polygonal figures, for instance, com-
plexity is defined as ' the number of indefinitely extended
straight lines which contain all the sides of the polygon'.
Order elements are analysed into the following: V, or ver-
tical symmetry, E, or equilibrium, R, or rotational sym-
metry ; HVy or relation to a horizontal-vertical network ; and
F, or unsatisfactory form, which is a kind of rag bag involv-
ing factors such as the following: too small distances from
vertices to vertices or to sides, or between parallel sides;
angles too near o° or 1800; other ambiguities; unsupported
re-entrant sides; diversity of niches; diversity of directions;
lack of symmetry. All the terms used in this formula are pre-
cisely defined, and it is possible for any given polygonal

328 Part Two • Personality and Social Life

figure to arrive at a measure M which, in terms of Birkhoff's
theory, should give its degree of aesthetic measure.

The crucial question, of course, is a factual one. Does the
formula actually work? We may agree or disagree with the
theoretical development, and we may like or dislike the
whole orientation of Birkhoff's work. If he should happen
to succeed in predicting aesthetic preference judgements on
the basis of a formula, however, we could hardly neglect
such an important contribution. The answer seems to be
that the formula does work up to a point, but that it does not
work very well. Agreement with actual expressed preference
judgements by large numbers of people is relatively slight.

There appear to be three main reasons for this. In the
first place, Birkhoff bases part of his theory on an assump-
tion which is not, in fact, borne out. In his discussion of the
complexity elements in aesthetic appreciation he links it
with the effort made by the observer in the act of percep-
tion, and he seems to equate this effort with the actual
amount of muscular exertion. This is what he says : ' Sup-
pose that we fix attention upon a complex polygonal tile.
The act of perception involved is so quickly performed as to
seem nearly instantaneous. The feeling of effort is almost
negligible while the eye follows the successive sides of the
polygon and the corresponding motor adjustments are
effected automatically. Nevertheless, according to the point
of view advanced above, there is a slight feeling of tension
attendant upon each adjustment, and the complexity C will
be measured by the number of sides of the polygon.' Now
even a very elementary knowledge of psychology would
have taught Birkhoff that in actual fact the eye, in perceiv-
ing any kind of figure, polygonal or otherwise, does not, in
fact, 'follow the successive sides'. By photographing actual
eye-movements during reading, during the perception of
pictures, of polygonal figures, and of other objects, it has
been established that the eye does not smoothly follow along
the lines constituting the percept, but that rather it rests for
a brief moment and then, by means of a so-called saccadic
movement, jumps to another point, resting there again for a

The Psychology of Aesthetics 329

fraction of a second before jumping off again. In viewing
BirkhofT's hypothetical polygonal tile, the direction of these
jumps would be quite irregular and would certainly not fol-
low the lines of the figure being observed. There is, conse-
quently, no possibility of deriving the complexity element
from such pseudo-physiological considerations as those em-
ployed by Birkhoff. (It is interesting to note that several
other mathematicians and philosophers, in attempting to
build up some coherent account of visual perception and
appreciation of beauty, have fallen into the same trap. The
notion that one is free to make any kind of postulation in
psychology seems to die hard. Much time and effort, which
at present is wasted, could be put to better effect if, before
making assumptions of this kind, the writers bothered to find
out what the known facts of the situation are !)

Where Birkhoff 's first error was a factual one, his second
appears to have been a theoretical one. He makes the aes-
thetic measure of an object equal to the number of order
elements, and inversely equal to the number of complexity
elements. This is a purely a priori notion ; Birkhoff does not
appear to have carried out any experimental work what-
soever to check the truth of his hypothesis, or the accuracy
of his aesthetic measure. In a long series of experiments, the
conclusion was forced upon the present writer that the
general formula was wrong, and that the aesthetic measure
was not the ratio of the order and complexity elements, but
their product. In other words, M = 0 X C, not O/C. This
conclusion is best illustrated, perhaps, by looking at the
polygon having the highest M value in BirkhofT's table,
namely, the square. This figure is not particularly liked by
the majority of subjects, for the precise reasons which caused
Birkhoff to give it such a high rating. The square has a low
order of complexity and a relatively high degree of order.
The majority verdict seems to be that such an object is
* boring', 'uninteresting', 'too regular', and 'uninspired'.
In other words, it possesses too little diversity and too much
order to appeal. The most preferred objects seem to be those
having a high degree of complexity and a high degree of

330 Part Two • Personality and Social Life

order. When formulae were constructed on this basis, agree-
ment with observed preference judgements rose to a respect-
able height, and in terms of this new and changed formula
it is possible to say that we can predict aesthetic reactions in
terms of a simple objective mathematical formula, thus appar-
ently confirming the notion expressed in the ancient rhyme :

For he that reades but Mathematicke rules
Shall finde conclusions that availe to worke
Wonders that passe the common sense of men.

The third reason why BirkhofF went wrong lies in his
failure to take into account some additional complexities. In
addition to the general preference judgements character-
istic of human beings in the group, we must take into ac-
count also considerable individual differences. These differ-
ences seem to relate particularly to preferences for order or
complexity elements respectively. In other words, some
people appear to prefer aesthetic objects having a relatively
high degree of complexity, whereas others prefer aesthetic
objects having a relatively high degree of order. I may per-
haps illustrate this from some experiments done with the
appreciation of poetry. Those with a strong liking for a pre-
ponderance of order elements tended to prefer poems with a
simple regular beat; a simple straightforward rhyming
scheme having a regular and heavily accentuated rhythm.
The following stanza may serve as an example of this :

Into the Silence of the empty night
I went, and took my scorned heart with me,
And all the thousand eyes of Heaven were bright;
But Sorrow came and led me back to thee.

By contrast, here is a stanza from a poem preferred by those
generally liking more complex types of phrase forms, i.e. an
unorthodox irregular rhyming scheme, and a much less
obvious rhythm:

Thou art not lovelier than lilacs - no,
Nor honeysuckle ; thou art not more fair
Than small white single poppies, - I can bear
Thy beauty.

The Psychology of Aesthetics 331

On the average, the two poems from which these stanzas
were taken would be liked equally well, but there would be
considerable disagreement between subject, those liking the
one disliking the other, and vice versa. There is some evi-
dence to suggest that in part, at least, these differences are
due to temperamental factors. Extraverts tend to prefer the
simple type of poem with a regular rhyming scheme, and the
heavily accentuated rhythm; introverts prefer the more
complex type of poem with the irregular rhyming scheme
and the less obvious rhythm.

In addition, there is some evidence to suggest that prac-
tice and familiarity also play a part in determining a per-
son's appreciation. It has been found, for instance, that
when subjects were asked to make preference judgements
for simple and complex chords struck on the piano, their
preferences tended to go towards the simple and more fami-
liar types of chords. However, when the experiment was
repeated a large number of times the simple familiar chords
began to lose their appeal and the more complex, unfamiliar
ones began to become more popular. Much the same has
been found with respect to other types of art objects, as in-
deed one might have suspected on common-sense grounds.
Apparently, familiarity itself must be counted as an addi-
tional order element in theformula,so that simple repetition,
of viewing, or hearing, may change the aesthetic measure of
a given object. These additional complications can all be
taken into account in the final formula, of course, but it can
be seen that the method is by no means as simple as sug-
gested by Birkhoff. Unfortunately, there is at present little
interest among psychologists in the experimental study of
aesthetics, and the very promising beginning made in the
fields of colour preferences and the aesthetic measure are not
likely to be followed up on a sufficient scale to make rapid
progress in this very difficult field likely.

Throughout this chapter we have been concerning our-
selves with formal aspects of art. These have always been of
major interest to psychologists because they alone lend
themselves easily to measurement and, hence, to the formu-

332 Part Two • Personality and Social Life

lation of laws and the accumulation of experimental evi-
dence so desirable when exact statements of relationship are
required. However, rightly or wrongly, the man in the
street, the literary critic, and the artist have usually shown
much more interest in a rather different kind of analysis.
This type of analysis deals with content rather than with
form. It is subjective rather than objective. It does not make
exact statements in a numerical form, but rather tries to
convey impressions by means of words. These features,
which render it somewhat suspect to the scientist, make it
much more readily acceptable to a wide variety of people
who are more interested in the humanities than in science,
and who do not look kindly upon any attempt to make
aesthetic experiences amenable to scientific laws.

Much of the kind of analysis I have in mind here has been
done in the fields of the novel and drama, perhaps largely
because, in these fields, content is very much more impor-
tant than form, just as scientific analysis has been most pro-
minent in the fields of visual arts and music, where form is
more important than content. Best known among the differ-
ent types of analysis of content is probably that carried out
by psychoanalysts, who have concentrated a considerable
amount of energy on this field. An example may show just
precisely what these writers are attempting to do, and we
can then go on to discuss how far what they are trying to do
is feasible and successful.

As we cannot discuss in detail all the many attempts that
have been made by psychoanalysts, we shall have to restrict
ourselves to one or two examples, and in order to be fair I
have chosen as my first exhibit what has been universally
considered to be the most successful of all these attempts. I
refer to Ernest Jones's attempt to explain what he calls
* Hamlet's mystery' in terms of the Oedipus complex. In this
he has but followed in the footsteps of Freud himself, who
made a similar suggestion in a footnote to one of his books.

What, then, is 'Hamlet's mystery'? This, according to
Jones, is the peculiar hesitancy shown by Hamlet in seeking
to obtain revenge for the murder of his father. Three types

The Psychology of Aesthetics 333

of hypothesis have been put forward to explain this 'Sphinx
of modern Literature'. The first of these hypotheses sees the
difficulty in the performance of the task in Hamlet's tem-
perament, which is not supposed to be suited to effective
action of any kind. This view, seeking to account for Ham-
let's inhibition by some difficulty in his constitution, was
originally advanced by Goethe, Schlegel, and Coleridge.
This view roughly maintained that because of his highly de-
veloped intellectual powers, and his broad and many-sided
sympathies, Hamlet never took a simple view of any prob-
lem, but always saw a number of different aspects and pos-
sible explanations. Hence, no particular course of action
ever seemed unequivocal and obvious, and therefore his
scepticism and reflective powers tended to paralyse his con-
duct in practical life. In terms of the concepts discussed in
detail in an earlier chapter, we might say that Hamlet was
a highly introverted person, probably with strong neurotic
tendencies - a person liable to lose himself in abstract trains
of thought at the expense of contact with reality.

The second hypothesis seeks for a cause, not in the person-
ality of Hamlet, but rather in the difficulty of the task itself.
There are many sub-varieties of this kind of hypothesis, but
they all agree in suggesting that, to depose a reigning sover-
eign without, at the same time, losing one's own life, was an
extremely difficult matter, particularly if, at the same time,
an attempt is to be made not only to slay the murderer but
also to convict him of his murder.

Jones rejects both these hypotheses as accounting for
Hamlet's vacillation, and goes on to say:

If this lies neither in his incapacity for action in general, nor in
the inordinate difficulty of the task in question, then it must of
necessity lie in the third possibility, namely in some special feature
of the task that renders it repugnant to him. This conclusion, that
Hamlet at heart does not want to carry out the task, seems so ob-
vious that it is hard to see how any critical reader of the play could
avoid making it.

What Jones has, in fact, suggested is that Hamlet had
some repugnance to carrying out the task of revenge, but he

334 Part Two ' Personality and Social Life

is handicapped in maintaining this view because nowhere in
the play is there any evidence in Hamlet's soliloquies of such
repugnance. This, however, presents no difficulty to Jones ;
if there is no evidence in Hamlet's words of any repug-
nance, of any cause for his inhibition, then the simple con-
clusion is that he himself was unaware of the nature of his
repugnance - in other words, that he was unconscious of it.
Having thus thrust the explanation of the mystery which he
is seeking into the unconscious, Jones goes on to try to find
out what it is that has been repressed. He makes a start by
stating a law which runs like this : ' That which is inaccept-
able to the herd becomes inacceptable to the individual
unit.' He goes on to say,

It is for this reason that moral, social, ethical, or religious in-
fluences are hardly ever 'repressed', for as the individual origin-
ally received them from his herd, they can never come into conflict
with the dicta of the latter. . . . The contrary is equally true, namely
that mental trends 'repressed' by the individual are those least
acceptable to his herd.

It would follow from these considerations that what has
been repressed, and what caused Hamlet to hesitate, is
something which is not acceptable to him or to his herd.

This 'something' Jones looks for by examining Hamlet's
attitude towards the object of his vengeance, Claudius, and
to the crimes which have to be avenged. These are first
Claudius's incest with the Queen, and second his murder
of Hamlet's father, Claudius's brother. In trying to explain
Hamlet's attitude towards Claudius, Jones claims that this
is not simply one of mere execration, but that there is a
complexity arising in the following way:

The uncle has not merely committed each crime, he has com-
mitted both crimes, a distinction of considerable importance, for the
combination of crimes allows the admittance of a new factor, pro-
duced by the possible inter-relation of the two, which prevents the
results from being simply one of summation. In addition it has to
be borne in mind that the perpetrator of the crimes is a relative,
and an exceedingly near relative.

The Psychology of Aesthetics 335

Having taken us this far, Jones now explains the mechan-
ism of the so-called Oedipus complex, which he believes will,
at the same time, explain Hamlet's hesitancy. Briefly, this
complex is supposed to arise because of a strong sexual link
between the boy child and his mother. (Girls are somehow
left out of this intriguing constellation, and although Freud
has, at times, made perfunctory efforts to fit them in, these
cannot be said to be anything like as striking as those leading
to the concept of the Oedipus complex.) In this liaison, the
father is experienced as a successful rival who comes be-
tween the child and his mother. Consequently, the boy
wants to kill the father and marry the mother, a feat actually
accomplished by Oedipus in the Greek legend and tragedy
of Sophocles. It is because of this similarity between the
story of Oedipus and the alleged desires of young boys that
this complex has received its name.

It is in terms of the Oedipus complex then that Jones
would explain Hamlet's mystery. Having in his youth re-
pressed his aggressive feelings towards his father, and his
sexual feelings towards the mother, he now receives news of
the father's death and the mother's second marriage. As a
consequence, Jones points out,

The long-' repressed ' desire to take his father's place in his
mother's affection is stimulated to unconscious activity by the sight
of some one usurping this place exactly as he himself had once
longed to do. More, this someone was a member of the same
family, so that the actual usurpation further resembled the imag-
inary one in being incestuous. Without his being at all aware of it
these ancient desires are ringing in his mind, are once more
struggling to find expression, and need such an expenditure of
energy again to 'repress' them that he is reduced to the deplor-
able mental state he himself so vividly depicts.

Add to all this the ghost's revelation that the uncle is also
the murderer of his father, and Hamlet's unconscious is
clearly in a considerable turmoil. According to Jones, his
attitude towards Claudius becomes a very complex one. He
detests his uncle, but it is the jealous detestation of one evil-
doer towards his successful fellow. This makes it difficult for

336 Part Two • Personality and Social Life

him to denounce his uncle, because the more vigorously he
does so, the more powerful does he stimulate the activity of
his own unconscious and 'repressed' complexes. This, then,
is Jones's explanation for Hamlet's failure. He cannot obey
the call of duty to slay his uncle because this is linked with
the call of his nature to slay his mother's husband, whether
this is the first or the second ; the latter call is strongly ' re-
pressed', and therefore necessarily the former also. In other
words, Hamlet, having as a young boy repressed his (hypo-
thetical) desire to slay his father, now, for no obvious reason,
transfers this repression to his mother's second husband and
therefore is prevented from carrying out his vengeance.

This explanation may or may not appeal to the reader,
but before discussing it, let us first of all try to get clear pre-
cisely what it is that is being said. Jones appears to suggest
that Hamlet is suffering from an Oedipus complex, which in
some way which is not very clear and seems to rest on a quite
arbitrary assumption, causes him to hesitate to kill Claudius.
But as a statement, of course, this has no meaning. Hamlet,
after all, is a non-existent person, a figment of the imagina-
tion, and to assert that a non-existent person is activated by
an Oedipus complex is not a very enlightening statement. It
clearly is impossible to refute or to support in any meaning-
ful way, but Jones, of course, only uses this statement as a
stepping-stone. He wants to inquire into the relation of
Hamlet's complex to the inner workings of Shakespeare's
mind. Thus, he goes on to say, ' It is here maintained that
this conflict is an echo of a similar one in Shakespeare him-
self, as to a greater or less extent it is in all men.' In other
words, Jones is asserting that Hamlet is a kind of Thematic
Apperception Test which can be used to diagnose Shake-
speare's neurosis. As nothing whatever is known of Shake-
speare's personality such an assertion, of course, can be
made quite confidently. It also can neither be proved or dis-
proved. In particular, any kind of disproof is ruled out be-
cause the conflict connected with the Oedipus complex is,
inevitably, an unconscious one, and therefore must for ever
remain hidden.

The Psychology of Aesthetics 337

But this is not the end of the story. If Shakespeare, uncon-
sciously and unwittingly, endowed Hamlet with uncon-
scious desires which make it impossible for him to carry out
his duty, he also produced a drama which appeals to us be-
cause the same conflict (unconsciously, of course) is present
in our own minds. Thus, Shakespeare's unconscious, through
the unconscious of Hamlet, is calling to the unconscious of
the various members of the theatre audiences who hear the
play and who unconsciously recognize and appreciate all
these unconscious determinants.

But the matter does not rest there either. Freud and Jones
both explicitly assert that the Oedipus complex is charac-
teristic of all human beings. In view of that, it is a little
difficult to see how the Oedipus complex can be adduced
to explain the actions of any particular person. This point
has often been mentioned in criticism of Freudian theories ;
by making their conceptions universal, i.e. by applying
them to all human beings, they also make it quite impossible
to use these conceptions in an explanatory manner. Only
that which affects different people differently can be made
into an explanatory principle to account for differences be-
tween them. If everyone has an Oedipus complex, then
the possession of one does not differentiate Hamlet from
Claudius or Laertes. Nor does it differentiate him from
Shakespeare, or any member of the audience during the
play.

Perhaps the diagnosis has to be more detailed still, but
how is that possible? Freud himself makes it clear that only
by detailed analytic sessions, extended over many years, and
assisted by procedures of dream interpretation and free asso-
ciation, can reliable diagnoses be made. To make such
a diagnosis of a person, 300 years dead, of whose life
practically nothing is known, on the strength of a very
esoteric interpretation of a few lines probably, but not cer-
tainly, written by that person, on the basis of a story already
existing at that time in several different forms, seems to be a
somewhat extravagant claim. Whether the reader feels that
the explanation given by Jones is a likely one or not, he will

338 Part Two • Personality and Social Life

agree that the whole process of arriving at this explanation
is a literary rather than a scientific device, and, while it un-
doubtedly is more colourful than the more sober scientific
types of research, it can hardly aspire to the same degree of
confidence in its conclusion.

Much the same may be said about a slightly less widely
known psychoanalysis conducted by Schilder on Alice in
Wonderland, and through her on Lewis Carroll. Schilder
bases himself on a saying by Freud, according to which non-
sense in dreams and so-called unconscious thinking signifies
contempt and sneering. He therefore says, ' We may expect
that nonsense literature is the expression of particularly
strong destructive tendencies of a very primitive character.'
Whence do these destructive tendencies come? Schilder
maintains that they arose because Carroll, being one of a
large family, never got the full love of his parents and there-
fore hated and wanted to destroy his various brothers and
sisters. Carroll, as Schilder points out, liked to play with
toads and snails and earth-worms, while Alice is in continu-
ous fear of being attacked or blamed by the animals. ' Do the
insects', asks Schilder, 'represent the many brothers and
sisters who must have provoked jealousy in Carroll?'

Somewhat inconsequentially, Schilder then goes on to
ask, with respect to Lewis Carroll, ' What was his relation to
his sex organ, anyhow?' The answer he gives relates to a
theory adumbrated by another well-known psychoanalyst,
Fenichel, who suggested the possibility that little girls might
become symbols for the phallus. In support of this view,
Schilder points out that Alice changes her form continually.
She is continually threatened and continually in danger.
Thus, apparently, Lewis Carroll is suffering from a castra-
tion complex, which thus appears as a strong motivating
factor in his writings.

What about the reader? Schilder declares that to him
Carroll appears to be a particularly destructive writer. He
asks whether such literature might not increase destructive
attitudes in children beyond the measure which is desirable.
Thus, Alice in Wonderland and Through the Looking Glass join

The Psychology of Aesthetics 339

the ranks of the horror comics, a somewhat curious juxta-
position!

This is by no means the only psychoanalytic interpreta-
tion of Alice\ others are available which make use of quite
different hypotheses and emerge with quite different conclu-
sions. The reader who has a fertile imagination and some
knowledge of the psychoanalytic system will find many hours
of unalloyed amusement by going through the adventures of
Alice, trying to link them up with the more obvious psycho-
analytic concepts which they may call to his mind. Such a
procedure may do little to throw light on the psychology of
aesthetics, but it will convince the reader of the unparalleled
fertility of analogies to be found in the Freudian system, and
of the utter impossibility of subjecting these analogies to any
rational test of truth or falsehood.

RECOMMENDED FOR FURTHER
READING

i. Weitzenhoffer, A. M. Hypnotism. An Objective Study in Sug-
gestibility. New York: John Wiley & Sons, Inc., 1953. (for
Chapter 1)

2. Inbau, F. E., and Reid, J. E. Lie Detection and Criminal Interro-
gation. Third Edition. Baltimore : The Williams & Wilkins Com-
pany, 1953. (for Chapter 2)

3. West, D. J. Psychical Research Today. London: Duckworth,
1954. (for Chapter 3)

4. Hall, C. S. The Meaning of Dreams. New York: Harper &
Bros., 1954. (for Chapter 4)

5. Eysenck, H. J. The Structure of Human Personality. London:
Methuen & Co. Ltd., 1953. (for Chapter 5)

6. Pavlov, I. P. Conditioned Reflexes. London: Oxford University
Press, 1927. (for Chapter 6)

7. Eysenck, H.J. The Psychology of Politics. London: Methuen &
Co., 1954. (for Chapter 7)

8. Birkhoff, G. D. Aesthetic Measure. Cambridge, Mass. : Harvard
University Press, 1933. (for Chapter 8)

INDEX

Adaptability, of the human body,

79

Aesthetic appreciation, Birkhoff

on, 326 fF.
psychology of, 308-9
qualitative aspect, 322
style preferences, 322, 323
Aggressive behaviour, 234, 235,
264, 279, 282, 288, 294, 299,
300, 301
analysis of, Fig. 9, p. 300
barriers to, 279
Aims, of the psychologist, 12
Aklaq-i-Jalali, quoted, 72-3
Albicerius, 109
Alcatraz, 18, 19
Alcohol, 103
Allport, G. W., 175
Amnesia, 31, 40, 41, 58, 65, 75,

104, 105, 254
Anaesthesia, hypnotic, 31, 37,

58, 254
Anaesthetics, 38, 39
Analgesia, 31, 37, 38, 66
Analysis, 204, 220, 223, 224,

227
Animal magnetism, 25, 26, 27,

28, 30, 60, 62, 63
Anti-Semitism, 283
Anti-social activities, 259, 260,
264, 294
induced by hypnosis, 52, 53,

54> 55, 56, 58
Anxiety, 261

Art, formal elements of, 308-25
formula for measurement of,

326, 327
Artemidorus, 146, 155
Association of ideas, 155, 156,

157, 158. !59, 170, Hi, 237.
278
law of, 271, 293
'Atomistic hypothesis', 317, 318

Attention, and distraction, pheno-
mena of, 240, 241
Autonomic responses, 78, 260, 261
Avicenna, 73, 75, 76

Babinski reflex, 49
Behaviour, animal, 231 ff., 248,
249, 250, 253, 260, 261,
271, 272

barriers in, 279 ff.

human, II, 12, 21, 58, 63, 180,
181, 198, 200, 207, 217, 253,
259, 279 ff.

laws governing, 217, 218

rating of, 184, 185

See also Sex
Bernheim, Professor, 28, 62, 64
Birkhoff, George D., 326 ff.
Blindness, congenital, 311, 312
Blood, circulation of, 75, 78
Blood pressure, 79, 84
Blood-stream, 77, 78, 79
Body, adaptability of, 79

in relation to the mind, 77,

106
Braid, James, 60, 61
Brain, activity of, 236, 238
'Brain waves' (electro-encephalo-
gram recordings), 65
Bramwell, J. M., quoted, 61-2
B.B.G., 13

Calculating machines, 205
Cannon, W. B., 79
Garington, W. W., 137
Catalepsy, in hypnosis, 58, 63
Central nervous system, dissocia-
tion of, 65

See also Nervous system
Charcot, J. M., 35, 63, 64
Childbirth, hypnosis in, 39
Children, socialization of, 293

training of, 20, 21, 22

34 1

342

Index

Clairvoyance, 106-41
Colour dimensions, 313 ff.

preferences, 313 ff.
Communists, 280-2, 288, 295-

301
Complexes, 171, 227
'Conditionability', 262, 263, 279,

293, 295, 297
Conditioned reflex, 65, 231 ff.,

241-46, 251, 252, 255
experiments in, 236, 237, 255 ff.
positive, 247
Conditioning, 257, 258, 260,
261, 262, 278
classical, 258, 260, 261
instrumental, 258, 259
Confessions, induced by lie-de-
tection techniques, 92, 93,
99, 100
Consciousness, 106, 151, 171
Control groups, in psychological

investigations, 170, 205
Controlled experiments, 67, 68,

69, 70
Conventions, respect for, 279,

280, 294
Coover, H., 121, 122
Criminal activity, in hypnotized
persons, 52, 53
protection against, 100, 10 1
Croesus, 109

Cultural knowledge, 320, 321
influence, 321

Dark vision, 178, 179, 180

Davey, S.J., in, 112, 113

Day-dreams, 172

Death, survival after, 106, 132

Deception, 84, 91, 92

Defoe, Daniel, quoted, 10 1, 102

Delphi, oracle at, 109, no

Description, meaning in scientific

usage, 233, 234
Determinism, 152
Diagnosis of neurotics, 194, 195,

204, 206
Disinhibition, 247, 252

Dissociation, as explanation of

hypnotic phenomena, 65
Dolorimeter, 238, 239, 240
Dols, 238
Dreams, 142-74

characters in, 142, 143, 144

coloured, 144

emotional content, 142, 144,

145, 162
experimental work in, 171
Freudian theory, 151 ff.
latent, 154, 156, 162, 170
manifest, 154, 156, 162, 170
mechanisms :

condensation, 161
displacement, 162, 163
dramatization, 159, 160
precognitive, 146, 150
primitive interpretation, 145
prophetic, 118, 135, 1458*.
settings of, 142, 143
symbolism in, 143, 144, 145,

155, 159, l6o> l6l> l63> l64

167, 168, 169, 173
Drugs, truth, 71-105
Du Hussey, Charles, quoted, 26,

27
Dunne, J. W., 146-9
'Dynamic' theory, 228
Dynamometer, 209, 210, 239,

240
Dysthymics, 254, 255, 257, 263,

264

Education, coaching, 15

equalization, 14

examination nerves, 15, 16

use of intelligence tests, 13 ff.
Effect, law of, 258, 259, 261
Ego, 153, 154
Elberfeld horses, 124, 125
El Capellan de la Virgin, 232
Electricity, reaction of skin to,
75,86,87,255,309

transmission, in scientific ex-
periments, 237
Electric shock, 251, 255

Index

343

Electro-encephalogram record-
ings, 65
Electronics, 77, 218
Elkisch, drawing test, 132, 133
Elliotson, John, 38
Emotion, James-Lange law, 76
Emotional responses, 76 fF., 218,

309
effect of hypnosis on, 47
index of, 75
instability of, 197, 204
involuntary, 76
in war-time, 79
Empiricism, 266 ff.
Esdaile, G., 38
Estabrooks, G. H., 125
Eugenics, 158

Excitation, 241, 244 ff., 253, 257,
262
and inhibition, 241
excessive, 255, 257
in ratio to inhibition, 253
physiological basis, 247
'Experimental neurosis', 250, 251
Experiments in hypnosis, 45, 46, 67

in psychology, 61
Experiment with Time, An (Dunne),

146
Explanation, meaning in scientific

usage, 233, 234
External stimuli, 77, 86, 87, 144,
145, 172, 237, 238, 239,
241 ff, 251, 255, 261
inhibiting, 239
positive, 240
'Extinction', internal inhibition

of, 245, 246
Extra-sensory perception, 106,
119, 121, 122, 123, 138
auditory cues in, 124
correlates of, 132
experiments in, 121 ff., 132,

I33> 134
Extraversion, 197, 198, 256, 257,

262, 263, 278, 279, 297

Extraverts, 257, 264, 299, 310,

323, 331

Eye, pupil of, 239, 240
photographed, 240
reflex action, 255
saccadic movement, 328, 329

Fascists, 280-2, 288, 289, 295 ff.
Fatigue, decrease under hypnosis,

45,46
Fear, 260, 261

during lie-detection tests, 87,

88,89
Fechner, G. T., 326
Fenichel, Otto, 338
France, political structure, 296
Freud, S., 35, 7 1 , 1 5 1 , 1 52, 1 53, 1 54,

155, 156, 158, 159, 173, 218,

262, 332, 337, 338
'Freudian abreaction', 52
Freudian mechanisms, 159 ff.
Freudian theory of hypnosis, 65,

66, 166, 168
Frink, M. F., 156, 161, 162

Gallup poll, 269, 272

Galton, Sir Francis, 158, 159

Gamblers, 107, 108

'Gestalt' school of psychology,

316, 317
Graphology, 227
Gravitation, law of, 233, 234,

235

Group therapy, 27, 28
'Guilt', knowledge of, 102

Habit, 244, 245

Hall, Calvin S., 164, 166, 167,

168, 173
Hallucinations, 31, 37

negative, 36, 37, 54, 58

positive, 36, 54, 58

post-hypnotic, 58
'Halo' effect, in rating human be-
haviour, 188
'Hamlet's mystery', 332 ff.
Handwriting analysis, 227
Hedonism, 271, 278, 293
Heredity, 14

344

Index

Hering, E., German psychologist,

236
Heymans, G., Brugmans, H., and

Wynberg, L., 123
Hodgson, Dr Richard, 116
Hornby, Sir Edmund, quoted,

114, 115
Hull, Clark L., quoted, 67
Human observation, unreliability

of, no ff.
Humphrey, G., 132, 133
Huxley, T. H., quoted, 106, 270
Hydrophobia, 75
Hypnodontics, 39
Hypnosis, 25-70, 106, 171, 172
categories, 58
chance elements in, 59
hypnoidal, 58
recent theories, 64, 65
regression, 47, 48, 49, 50, 51, 52
suggestibility in, 215
transcendence of, normal work-
ing capacity, 44, 45
See also Trance.
Hypnotic phenomena, 30, 31,

34, 60, 63, 67
'Hypnotic scales', 56, 57, 58

Eysenck-Furneaux scale, 57
Hypnotists, capabilities, 59
methods used by, 59
personal relationships, 59, 65
Hysteria, in relation to hypnosis,

64
Hysterics, 253, 254, 255, 257, 263

Id, 153, J54
Ideologies, 286, 287
Ideo-motor action, theory of, 66
'Implication', 204, 283, 284, 256,

287
Impulses, 78, 143, 244

barriers to satisfaction of, 279
Inbau, G., quoted, 91, 92, 95> 9&
Industrial psychologists, 191, 192
Inferiority, feelings of, 195
Inhibition, 66, 238, 241, 245-54,
262

Inhibition, conditional, 251, 252
differential, 248, 249
dissipation, 246, 247, Fig. 5b,

P- 249
excessive, 255, 257
external, 245
'extinction', 245-9, Fig. 5a, p.

248, 253
internal, 245
of delay, 252
physiological basis, 247
secondary, 251
with reinforcement, 252, 253
Insight, relevance to rating, 187,

188, 200
Insomnia, 75
Insusceptibility, 58
Intelligence quotients, of social

classes, 273
Intelligence tests, 13, 14, 15, 16,

194,211,212
Interreaction of personalities, 12,

187
Introversion, 197, 198, 256, 257,

262, 278, 297, 298
Introverts, 263, 279, 310, 323,

33i
Inventory, of personality, 193,

194, 198, 205

political, 288, 289, 296

social, 295, 302 ff.

Involuntary activities, 258

James, William, 76, 115, 116
James-Lange law of emotion, 76
Janet, Pierre, 253
Jews, attitude to, 283, 284
Jones, Ernest, 332 ff.
Jordan, F., 257

Jung, G. G., 59, 168, 173, i97»
198, 256, 257

Kant, on 'personality', 1 75
Kinsey, A. G., 294

Lange, G., Norwegian physiolo-
gist, 76

Index

Laubscher, Dr J. F., no
Law of effect, 258
of learning, 278
'Laws of composition,' 309, 310,

318
Laws of science, 269, 270
Lethargy induced by hypnosis, 63
Liebeault, A. A., 61, 62, 64
Lie detection, 71-105, 218
principle of, 77, 78
tests, 82, 83, 84, 85, 89

admissibility as legal evi-
dence, 95, 96, 97, 98
controlled response to, 89,

90

difficulties in, 81
interpretation of, 84, 85, 87
reliability of, 91, 95, 96
unresponsiveness to, 89
value of, 90, 91, 92, 93
validity of, 87
'Lie-scale', 203
Light-waves, 312, 313
Literature, psychological, on

bringing up children, 20
Locke, John, quoted, 268
Lodge, Sir Oliver, 116, 117
Love, classification by Avicenna,

75
Luria, A. R., Russian psychologist,

171

Mania, 75

Maudsley questionnaire, 195-7

202, 205
Meaning, doctrine of, 151, 152,

153
Measurement of personality, 175-

230
Medical treatment, hypnosis in,

38, 39. 60
Mediums, 107, 109, m, 112,

113, 115, 116, 118
Melancholia, 75
Memory, 106, 152, 156

experiments in regression, 47-

8, 49, 50, 51

345
under

Memory, improvement

hypnosis, 47, 48
Mental activity, 172, 173
Mental disorders, 64, 75, 89, 173,

253
and lie detection, 89
'Mentalistic explanation,' 235,

236, 240

Mesmer, Franz Anton, 25, 26, 27,

28, 29, 30, 31, 60
Mesmerism, 27, 29, 30, 52, 69, 70,

254
Methodology, scientific, 231
Mill, John Stuart, 67-8
Misconceptions of psychological

topics, 11, 12, 25
Moll, L., quoted, 28, 29
Motivation, effect of hypnosis on,

47, 59
Movement, voluntary, 78

involuntary, 76, 78
Muscles, work of, 78

'Negative Induction', law of, 236,

237, 238, 240, 241
Nervous disorders, treatment by

hypnosis, 26, 27, 31, 62
in animals, 250
Nervous impulses, 66, 244
Nervous system, autonomic, 78,
105, 258
central, 78, 179, 244, 247,

253, 319, 324
Neurons, 244
Neuroticism, 27, 43, 68, 69, 103,

104, 195, 198, 205
Neurotics, 68, 69, 105, 179, 186,

194, J95, 197, 201, 202, 203,

221, 224, 250, 253 ff.
Night-blindness, 178, 179, 180

Objective tests, 207 ff., 219
Objectivity, 312 ff.
Oedipus complex, 332 ff.
Oneirocritics, 146
Operation, surgical, under hyp-
nosis, 38

346 Index

Opinions, 285, 286

statements, 289 ff., 304 ff.
Oxygen consumption, 79

Pain, measurement of, 238, 239,
240, 260
suppression, by hypnosis, 38, 39
See also Analgesia.
Paralysis, 254

Parapsychology, 126, 131, 134,
137, 140, 141
displacement effects, 136, 137
salience, 138
Pascal, Blaise, 120
Pavlov, 65, 231, 232, 236, 241,
244, 245, 248, 250, 252, 253,
254, 255, 2575 264, 271
Pearce, Hubert, experiment on,

130* 131
Pearson, Karl, quoted, 159
Penfield, W., neuro-surgeon, 152
Pentothal, 103, 105
Perception of beauty, 308 ff.
Perceptual distinctions, 254
Persistence, measurement of,

207 ff.
Personality, 58, 199, 231-64
and parapsychology, 140
and politics, 265-307
characterization, 223, 224, 225
definition of, 175, 176
differences, 241
Freudian theory, 153, 154
in dreams, 144
inventory, 193, 194, 198, 205
measurement of, 175-230, 264

matching method, 225
non- cognitive, 132
origin of word, 1 76
rating, 184 ff., 192, 193, 204,

205, 206, 207, 216
role in psychology, 178, 179,

180, 181
study of, 179, 180, 181, 182
traits, 134, 207, 208 ff., 217,
256, 265, 310
neurotic, 179

Phrenology, 61

Physiological reactions, 76, 77, 78,
79, 81, 83, 84, 86, 87, 88,

309

measurement of, 218

under hypnosis, 65
Physiologists, 178, 179, 231
Physiology, 63, 77, 179

segmental analysis, 179
Piper, Mrs, 115, 116, 117
Plato, on beauty, 325
Pneumograph, 82
Police interrogation, lie-detection
tests as aid to, 91, 93, 94,

95
truth drugs, 105
Politics, 13, 265-307

differentiation between parties,

295, 296
relationships between parties,
280 ff., Fig. 6, p. 281, 288
Poltergeists, 106, 107
Polygraph, 81, 82, 93
Post-hypnotic amnesia, 65
Post-hypnotic suggestion, 31, 41,

42,43, 53, 58, 171
as curative agent, 43, 44
rationalization of, 42
See also Hypnosis; Suggesti-
bility.

Pratt, Dr, 131

Precognition, 135, 136, 137, 138,
146

Prediction, 135

Premonitions, 106

Prisons, 18, 19

'Projective techniques', 218, 219,
221, 222, 225, 226, 227, 228,
229

Psychiatrists, 183, 186, 190, 191,
204, 205, 206, 221, 225,

257
Psychiatry, 176, 177, 183
Psychical phenomena, 106 ff., 120,

121
contribution of psychology to

study of, 118

Index

347

Psychical research, 106, 108, 119,
121 ff.
Society for, 106, no, 114

American, 116
statistical basis, 119, 120
Psychoanalysis, 68, 69, 224
Psychoanalysts, 27, 39, 71, 170,

171, 190, 191,332
Psychogalvanic reflex, 86, 87, 88,

9i, 239, 255
See also Reflex actions
Psycho-kinesis, 107, 108, 139, 140
Psychological Types (Jung), 197
Psychologists, 164, 166, 171, 178,
183, 190, 191, 198, 200, 204,
205, 211, 225, 231, 257, 258,
286, 287, 308 ff.
aims of, 12

physiological, 178, 180
social, 178, 180, 181
Psychology, 176, 177, 178, 181,
191, 270
departments of, 12, 13
misconceptions surrounding, n,

12, 25
research in, 19, 20, 231 ff.
theories, 231 ff., 316, 317
Psychopaths, 263, 264
Psycho-physiology, 218
Psychosomatic disorders, 71 ff.
Psychosomatic relationships, 218
Psychotherapists, 69
Psychotherapy, 68, 69

hypnotic regression in, 47, 48,

49,50,5^52
Psychotic patients, 224
Pupillary reflex, 239
Puysegur, Marquis de, 31, 40

Quanun, quoted, 75-6
Questioning, 83, 84, 85
Questionnaires, 193, 194, 195,
197, 198, 199, 205, 206, 207,
216
faked answers, 203
interpretation of, 200, 201, 202,
203

Questionnaires, Maudsley, 195-6,
197, 202, 205
modern construction of, 204
political, 288
statistical analysis, 204
use of 'lie-scale', 203
value of, 199

Rapport, 39, 40

Rating procedures, 184 ff., 192,
193, 204, 205, 206, 207, 216
ability in, 188, 189
Rationalization, 42

See also post-hypnotic suggestion
Reassociation of ideas, 156
Recidivism, 18

'Recovery', phenomenon of, 246
Reflex actions, 65, 86, 87, 231-7,
251-61
Babinski, 49

conditioned, 21, 180, 241-5,
258, 278, 279
range of, 242, 243
secondary, 242, 243
psychogalvanic, 86, 87, 88, 91,

239
pupillary, 239
trace, 242, 243
Regression, hypnotic, 47, 48, 49,
50
used for psychotherapeutic pur-
poses, 51, 52
Research, psychical, 106, 108,
no, 114, 116, 119, 120, 121
scientific, 108, 231 ff.
Respiration, as index of emotion,

75
suppression in, 84
Response patterns, 84, 85, 220,

221
Rhazis, 72-3
Rhine, J. B., 125, 126, 127, 129,

130, 131, 135, 139, 140
Rorschach test, 133, 134, 220, 221,

222, 224, 225, 226, 228, 229
Rousseau, Jean Jacques, quoted,

152, 153

34^

Index

St Augustine, 109

Saint-Denis, Marquis Harvey de,

160
Schilder, on Alice in Wonderland,

338, 339
Schmeidler, 133, 134
Sciences, biological, 176, 178
social, 12, 13, 16, 17, 176, 177,

178
Scientific laws, 234
Scientific terms, discussion of,

233 ff., 240
Scientists, layman's view of, 108
Seances, 111, 112, 113, 117, 118
Sex, sexual behaviour, 155, 164,

165, 166, 259, 279, 288, 294
symbols, 155, 163, 164, 166,

167, 168, 169, 173
Shaw, G. B., 231, 232, 251, 262
Skinner, B. F., American psycho-
logist, 258
Sleep, as opposed to hypnotic

trance, 64, 65
Soal, S. G., 122, 128, 135, 136, 137
Social attitudes, 181, 279, 282,

283, 284, 286, 288
inventory, 289 ff., 302 ff.
Social classes, 272, 273, 274, 293,

294
in U.S.A., 278

See also Social status
Socialization, 259, 260, 262, 263,

264, 279, 293, 294
Social justice, and lie detection,

100, 101, 102
Social problems, 22
Social sciences, expenditure on,
12, 13, 16, 17
research, 265 ff.
Social status, 272 ff., 294

in relation to social class, 274,

275

in relation to politics, 276, 277

in U.S.A., 278
See also Social classes
Society for Psychical Research, 106
aims, 106

Socrates, quoted, 153, 154
Somnambulism, 63
Statistics, 107, 119, 120, 121, 183,
204, 205, 284, 285

fallacious use of, 17, 18

popular reaction to, 1 7

visual, 320
Stern, W., on personality, 175
Stimuli, 'complex', 310

inhibiting, 238, 239

positive, 236, 237, 240

'simple', 310

unconditional, 237, 241

See also External stimuli
Stimulus generalization, law of,

243, 244
Subjectivity, 308, 312 ff.
Suggestibility, 34, 35, 40, 41, 54,
55> 56, 60, 62, 65, 186

definition of, 212

primary, 215

secondary, 215

tertiary, 216

tests of, 212-15, 216

See also Hypnotic phenomena
Super-ego, 151, 153, 154, 262, 263
Susceptibility, 59, 60, 65
Sweat-glands, stimulation of, 86
Symbolism, 143, 144, 145, 155,
163, 164, 166, 167, 168, 169,

173

effi-

94,

Tautophone tests, 226
Teaching, factors affecting

ciency of, 14, 15
Techniques, experiments in,

95
in hypnotism, 59
in lie-detection, 83, 84, 85
See also Hypnosis; Lie-detection
Teeth, extraction under hypnosis,

39
Telepathy, 106-41

Television, effects of, 19
Thematic Apperception test, 219,

226, 299, 301, 302
'Third degree' practice, 99, 100

Index

349

Thompson, Jerry, 89, 90, 100, 101
Thought processes, 172, 173
'Total personality', 218
Trace reflex, 242, 252
Traits, in personality investiga-
tion, 134, 179, 207, 208, 212,
256, 265

measurement of, 208 ff.

See also Personality
Trance, hypnotic, 31, 32, 33, 35,
40, 60, 63, 64

awakening from, 34

deep, 58, 59

light, 58

medium, 58

susceptibility, 56, 57, 58
Transference, Freudian concept,

39, 64, 65
Truth drugs, 71-105
Tyrrell, G. N. M., 135, 136, 138

Unconditional stimuli, 237, 241,
255, 261

United States of America, army
recruits, 194, 195, 206
politics in, 266 ff., 282
social sciences, research in,
265 ff.
Units of pain, 238
U.S.S.R., Communists of, 282

Vega, Lope de, quoted, 232
'Verbal summator' test, 226
Visual art, 308 ff.
Voting behaviour, 181, 265, 276 ff.

See also Behaviour, human
Voluntary activities, 258

Wallace, A. R., 113
Water-divining, 106
Windelband, on personality, 1 75
Wish fulfilment, 151, 153, 161
Witch-doctors, 74-5, 81, no
Woodworth, R. S., American psy-
chologist, 195, 197, 198
Wundt, W., 158

More Pelicans by

H. J. Eysenck are described

overleaf

USES AND ABUSES OF PSYCHOLOGY

In this book H. J. Eysenck indicates both to what extent the claims
made for psychology are justified, and to what extent they fail to
have any factual basis. The discussion is very fully documented by
references to the most important and relevant researches carried out
in this country and abroad. Topics dealt with are the testing of
intelligence, selection procedures in schools and universities, voca-
tional guidance and occupational selection, psychotherapy and its
effects, national differences, racial intolerance, Gallup surveys, in-
dustrial productivity, and many others.

KNOW YOUR OWN I.Q..

This is at present the only book which permits the reader to deter-
mine his own I.Q.. In the first part of it the author describes clearly
what an I.O . is, how it can be applied, and what the shortcomings
of this system of rating may be. The second part of the book contains
eight sets of forty I.Q,. problems each. There are tables for converting
results into an I.Q,. rating, and explanations of the problems, to-
gether with the right answers, at the end of the book.

CHECK YOUR OWN I.O.

Know Tour Own I.Q. was a best-seller. It also provoked a mass of
critical comment and correspondence. Some readers claimed that
their answers were as good as the given solution, while others com-
plained that the book was a test of knowledge as much as intelligence.
A few protested that it was all too easy.

In this sequel, Check Tour Own I.Q,., Professor Eysenck answers these
criticisms fully and provides five new tests of the standard (omnibus)
type as a check. He has also added three specific tests which are
designed to sort out whether the reader shows more ability in verbal,
numerical, or visual-spatial terms.

For those with I.Q.'s of over 150 - to sharpen their wits or call
their bluff - he has provided a few sets of problems under the title
'Limbering Up for Intellectual Giants'.

- I: % ■' ':

Due

COLLEGE LIBRARY

Date Due
Returned Due

Returned

DEC 1 5 '65 1

Cf

MAY 2 3 •&

fc W*4 it

,11)1 3 1 1195

JUL 19 W95

UNIVERSITY OF FLORIDA

3 1262 04970 8612

/SO